Artificial Intelligence Nanodegree

Computer Vision Capstone

Project: Facial Keypoint Detection

---

Welcome to the final Computer Vision project in the Artificial Intelligence Nanodegree program!

In this project, you’ll combine your knowledge of computer vision techniques and deep learning to build and end-to-end facial keypoint recognition system! Facial keypoints include points around the eyes, nose, and mouth on any face and are used in many applications, from facial tracking to emotion recognition.

There are three main parts to this project:

Part 1 : Investigating OpenCV, pre-processing, and face detection

Part 2 : Training a Convolutional Neural Network (CNN) to detect facial keypoints

Part 3 : Putting parts 1 and 2 together to identify facial keypoints on any image!

---

*Here's what you need to know to complete the project:

1. In this notebook, some template code has already been provided for you, and you will need to implement additional functionality to successfully complete this project. You will not need to modify the included code beyond what is requested.

   a. Sections that begin with '(IMPLEMENTATION)' in the header indicate that the following block of code will require additional functionality which you must provide. Instructions will be provided for each section, and the specifics of the implementation are marked in the code block with a 'TODO' statement. Please be sure to read the instructions carefully!

1. In addition to implementing code, there will be questions that you must answer which relate to the project and your implementation.

   a. Each section where you will answer a question is preceded by a 'Question X' header.

   b. Carefully read each question and provide thorough answers in the following text boxes that begin with 'Answer:'.

Note: Code and Markdown cells can be executed using the Shift + Enter keyboard shortcut. Markdown cells can be edited by double-clicking the cell to enter edit mode.

The rubric contains optional suggestions for enhancing the project beyond the minimum requirements. If you decide to pursue the "(Optional)" sections, you should include the code in this IPython notebook.

Your project submission will be evaluated based on your answers to each of the questions and the code implementations you provide.

Steps to Complete the Project

Each part of the notebook is further broken down into separate steps. Feel free to use the links below to navigate the notebook.

In this project you will get to explore a few of the many computer vision algorithms built into the OpenCV library. This expansive computer vision library is now almost 20 years old and still growing!

The project itself is broken down into three large parts, then even further into separate steps. Make sure to read through each step, and complete any sections that begin with '(IMPLEMENTATION)' in the header; these implementation sections may contain multiple TODOs that will be marked in code. For convenience, we provide links to each of these steps below.

Part 1 : Investigating OpenCV, pre-processing, and face detection

• Step 0: Detect Faces Using a Haar Cascade Classifier
• Step 1: Add Eye Detection
• Step 2: De-noise an Image for Better Face Detection
• Step 3: Blur an Image and Perform Edge Detection
• Step 4: Automatically Hide the Identity of an Individual

Part 2 : Training a Convolutional Neural Network (CNN) to detect facial keypoints

• Step 5: Create a CNN to Recognize Facial Keypoints
• Step 6: Compile and Train the Model
• Step 7: Visualize the Loss and Answer Questions

Part 3 : Putting parts 1 and 2 together to identify facial keypoints on any image!

• Step 8: Build a Robust Facial Keypoints Detector (Complete the CV Pipeline)

---

Step 0: Detect Faces Using a Haar Cascade Classifier

Have you ever wondered how Facebook automatically tags images with your friends' faces? Or how high-end cameras automatically find and focus on a certain person's face? Applications like these depend heavily on the machine learning task known as face detection - which is the task of automatically finding faces in images containing people.

At its root face detection is a classification problem - that is a problem of distinguishing between distinct classes of things. With face detection these distinct classes are 1) images of human faces and 2) everything else.

We use OpenCV's implementation of Haar feature-based cascade classifiers to detect human faces in images. OpenCV provides many pre-trained face detectors, stored as XML files on github. We have downloaded one of these detectors and stored it in the detector_architectures directory.

import os
os.sys.path

['',
 '/home/ubuntu/anaconda3/envs/aind-cv/lib/python35.zip',
 '/home/ubuntu/anaconda3/envs/aind-cv/lib/python3.5',
 '/home/ubuntu/anaconda3/envs/aind-cv/lib/python3.5/plat-linux',
 '/home/ubuntu/anaconda3/envs/aind-cv/lib/python3.5/lib-dynload',
 '/home/ubuntu/anaconda3/envs/aind-cv/lib/python3.5/site-packages',
 '/home/ubuntu/anaconda3/envs/aind-cv/lib/python3.5/site-packages/IPython/extensions',
 '/home/ubuntu/.ipython']

Import Resources

In the next python cell, we load in the required libraries for this section of the project.

# Import required libraries for this section

%matplotlib inline

import numpy as np
import matplotlib.pyplot as plt
import math
import cv2                     # OpenCV library for computer vision
from PIL import Image
import time 

Next, we load in and display a test image for performing face detection.

Note: by default OpenCV assumes the ordering of our image's color channels are Blue, then Green, then Red. This is slightly out of order with most image types we'll use in these experiments, whose color channels are ordered Red, then Green, then Blue. In order to switch the Blue and Red channels of our test image around we will use OpenCV's cvtColor function, which you can read more about by checking out some of its documentation located here. This is a general utility function that can do other transformations too like converting a color image to grayscale, and transforming a standard color image to HSV color space.

# Load in color image for face detection
image = cv2.imread('images/test_image_1.jpg')

# Convert the image to RGB colorspace
image = cv2.cvtColor(image, cv2.COLOR_BGR2RGB)

# Plot our image using subplots to specify a size and title
fig = plt.figure(figsize = (8,8))
ax1 = fig.add_subplot(111)
ax1.set_xticks([])
ax1.set_yticks([])

ax1.set_title('Original Image')
ax1.imshow(image)

<matplotlib.image.AxesImage at 0x7f9e368071d0>

There are a lot of people - and faces - in this picture. 13 faces to be exact! In the next code cell, we demonstrate how to use a Haar Cascade classifier to detect all the faces in this test image.

This face detector uses information about patterns of intensity in an image to reliably detect faces under varying light conditions. So, to use this face detector, we'll first convert the image from color to grayscale.

Then, we load in the fully trained architecture of the face detector -- found in the file haarcascade_frontalface_default.xml - and use it on our image to find faces!

To learn more about the parameters of the detector see this post.

# Convert the RGB  image to grayscale
gray = cv2.cvtColor(image, cv2.COLOR_RGB2GRAY)

# Extract the pre-trained face detector from an xml file
face_cascade = cv2.CascadeClassifier('detector_architectures/haarcascade_frontalface_default.xml')

# Detect the faces in image
faces = face_cascade.detectMultiScale(gray, 4, 6)

# Print the number of faces detected in the image
print('Number of faces detected:', len(faces))

# Make a copy of the orginal image to draw face detections on
image_with_detections = np.copy(image)

# Get the bounding box for each detected face
for (x,y,w,h) in faces:
    # Add a red bounding box to the detections image
    cv2.rectangle(image_with_detections, (x,y), (x+w,y+h), (255,0,0), 3)
    

# Display the image with the detections
fig = plt.figure(figsize = (10,10))
ax1 = fig.add_subplot(111)
ax1.set_xticks([])
ax1.set_yticks([])

ax1.set_title('Image with Face Detections')
ax1.imshow(image_with_detections)

Number of faces detected: 13

<matplotlib.image.AxesImage at 0x7f9e35f13278>

In the above code, faces is a numpy array of detected faces, where each row corresponds to a detected face. Each detected face is a 1D array with four entries that specifies the bounding box of the detected face. The first two entries in the array (extracted in the above code as x and y) specify the horizontal and vertical positions of the top left corner of the bounding box. The last two entries in the array (extracted here as w and h) specify the width and height of the box.

---

Step 1: Add Eye Detections

There are other pre-trained detectors available that use a Haar Cascade Classifier - including full human body detectors, license plate detectors, and more. A full list of the pre-trained architectures can be found here.

To test your eye detector, we'll first read in a new test image with just a single face.

# Load in color image for face detection
image = cv2.imread('images/james.jpg')

# Convert the image to RGB colorspace
image = cv2.cvtColor(image, cv2.COLOR_BGR2RGB)

# Plot the RGB image
fig = plt.figure(figsize = (6,6))
ax1 = fig.add_subplot(111)
ax1.set_xticks([])
ax1.set_yticks([])

ax1.set_title('Original Image')
ax1.imshow(image)

<matplotlib.image.AxesImage at 0x7f9e35eed240>

Notice that even though the image is a black and white image, we have read it in as a color image and so it will still need to be converted to grayscale in order to perform the most accurate face detection.

So, the next steps will be to convert this image to grayscale, then load OpenCV's face detector and run it with parameters that detect this face accurately.

# Convert the RGB  image to grayscale
gray = cv2.cvtColor(image, cv2.COLOR_RGB2GRAY)

# Extract the pre-trained face detector from an xml file
face_cascade = cv2.CascadeClassifier('detector_architectures/haarcascade_frontalface_default.xml')

# Detect the faces in image
faces = face_cascade.detectMultiScale(gray, 1.25, 6)

# Print the number of faces detected in the image
print('Number of faces detected:', len(faces))

# Make a copy of the orginal image to draw face detections on
image_with_detections = np.copy(image)

# Get the bounding box for each detected face
for (x,y,w,h) in faces:
    # Add a red bounding box to the detections image
    cv2.rectangle(image_with_detections, (x,y), (x+w,y+h), (255,0,0), 3)
    

# Display the image with the detections
fig = plt.figure(figsize = (6,6))
ax1 = fig.add_subplot(111)
ax1.set_xticks([])
ax1.set_yticks([])

ax1.set_title('Image with Face Detection')
ax1.imshow(image_with_detections)

Number of faces detected: 1

<matplotlib.image.AxesImage at 0x7f9e35ec3a90>

(IMPLEMENTATION) Add an eye detector to the current face detection setup.

A Haar-cascade eye detector can be included in the same way that the face detector was and, in this first task, it will be your job to do just this.

To set up an eye detector, use the stored parameters of the eye cascade detector, called haarcascade_eye.xml, located in the detector_architectures subdirectory. In the next code cell, create your eye detector and store its detections.

A few notes before you get started:

First, make sure to give your loaded eye detector the variable name

eye_cascade

and give the list of eye regions you detect the variable name

eyes

Second, since we've already run the face detector over this image, you should only search for eyes within the rectangular face regions detected in faces. This will minimize false detections.

Lastly, once you've run your eye detector over the facial detection region, you should display the RGB image with both the face detection boxes (in red) and your eye detections (in green) to verify that everything works as expected.

# Make a copy of the original image to plot rectangle detections
image_with_detections = np.copy(image)   

# Loop over the detections and draw their corresponding face detection boxes
for (x,y,w,h) in faces:
    cv2.rectangle(image_with_detections, (x,y), (x+w,y+h),(255,0,0), 3)  
    
# Do not change the code above this comment!

# Extract the pre-trained eye detector from an xml file
eye_cascade = cv2.CascadeClassifier('detector_architectures/haarcascade_eye.xml')

# Detect the faces in image
eyes = eye_cascade.detectMultiScale(gray, 1.10, 6)

# Loop over the eye detections and draw their corresponding boxes in green on image_with_detections
# Get the bounding box for each detected eye
for (x,y,w,h) in eyes:
    # Add a red bounding box to the detections image
    cv2.rectangle(image_with_detections, (x,y), (x+w,y+h), (0,200,0), 2)

# Plot the image with both faces and eyes detected
fig = plt.figure(figsize = (6,6))
ax1 = fig.add_subplot(111)
ax1.set_xticks([])
ax1.set_yticks([])

ax1.set_title('Image with Face and Eye Detection')
ax1.imshow(image_with_detections)

<matplotlib.image.AxesImage at 0x7f9e35e21ba8>

(Optional) Add face and eye detection to your laptop camera

It's time to kick it up a notch, and add face and eye detection to your laptop's camera! Afterwards, you'll be able to show off your creation like in the gif shown below - made with a completed version of the code!

Notice that not all of the detections here are perfect - and your result need not be perfect either. You should spend a small amount of time tuning the parameters of your detectors to get reasonable results, but don't hold out for perfection. If we wanted perfection we'd need to spend a ton of time tuning the parameters of each detector, cleaning up the input image frames, etc. You can think of this as more of a rapid prototype.

The next cell contains code for a wrapper function called laptop_camera_face_eye_detector that, when called, will activate your laptop's camera. You will place the relevant face and eye detection code in this wrapper function to implement face/eye detection and mark those detections on each image frame that your camera captures.

Before adding anything to the function, you can run it to get an idea of how it works - a small window should pop up showing you the live feed from your camera; you can press any key to close this window.

Note: Mac users may find that activating this function kills the kernel of their notebook every once in a while. If this happens to you, just restart your notebook's kernel, activate cell(s) containing any crucial import statements, and you'll be good to go!

### Add face and eye detection to this laptop camera function 
# Make sure to draw out all faces/eyes found in each frame on the shown video feed

import cv2
import time 

# Extract the pre-trained face detector from an xml file
face_cascade = cv2.CascadeClassifier('detector_architectures/haarcascade_frontalface_default.xml')

# wrapper function for face/eye detection with your laptop camera
def laptop_camera_go():
    # Create instance of video capturer
    #cv2.namedWindow("face detection activated")
    vc = cv2.VideoCapture(0)

    # Try to get the first frame
    if vc.isOpened(): 
        rval, frame = vc.read()
    else:
        rval = False
    
    # Keep the video stream open
    while rval:
        # Convert the RGB  image to grayscale
        gray = cv2.cvtColor(frame, cv2.COLOR_RGB2GRAY)

        # Detect the faces in image
        faces = face_cascade.detectMultiScale(gray, 1.25, 6)

        # Make a copy of the orginal image to draw face detections on
        image_with_detections = np.copy(frame)

        # Get the bounding box for each detected face
        for (x,y,w,h) in faces:
            # Add a red bounding box to the detections image
            cv2.rectangle(image_with_detections, (x,y), (x+w,y+h), (255,0,0), 3)
        
        # Plot the image from camera with all the face and eye detections marked
        cv2.imshow("face detection activated", frame)
        
        # Exit functionality - press any key to exit laptop video
        key = cv2.waitKey(20)
        if key > 0: # Exit by pressing any key
            # Destroy windows 
            cv2.destroyAllWindows()
            
            # Make sure window closes on OSx
            for i in range (1,5):
                cv2.waitKey(1)
            return
        
        # Read next frame
        time.sleep(0.05)             # control framerate for computation - default 20 frames per sec
        rval, frame = vc.read()    

# Call the laptop camera face/eye detector function above
laptop_camera_go()

---

Step 2: De-noise an Image for Better Face Detection

Image quality is an important aspect of any computer vision task. Typically, when creating a set of images to train a deep learning network, significant care is taken to ensure that training images are free of visual noise or artifacts that hinder object detection. While computer vision algorithms - like a face detector - are typically trained on 'nice' data such as this, new test data doesn't always look so nice!

When applying a trained computer vision algorithm to a new piece of test data one often cleans it up first before feeding it in. This sort of cleaning - referred to as pre-processing - can include a number of cleaning phases like blurring, de-noising, color transformations, etc., and many of these tasks can be accomplished using OpenCV.

In this short subsection we explore OpenCV's noise-removal functionality to see how we can clean up a noisy image, which we then feed into our trained face detector.

Create a noisy image to work with

In the next cell, we create an artificial noisy version of the previous multi-face image. This is a little exaggerated - we don't typically get images that are this noisy - but image noise, or 'grainy-ness' in a digitial image - is a fairly common phenomenon.

# Load in the multi-face test image again
image = cv2.imread('images/test_image_1.jpg')

# Convert the image copy to RGB colorspace
image = cv2.cvtColor(image, cv2.COLOR_BGR2RGB)

# Make an array copy of this image
image_with_noise = np.asarray(image)

# Create noise - here we add noise sampled randomly from a Gaussian distribution: a common model for noise
noise_level = 40
noise = np.random.randn(image.shape[0],image.shape[1],image.shape[2])*noise_level

# Add this noise to the array image copy
image_with_noise = image_with_noise + noise

# Convert back to uint8 format
image_with_noise = np.asarray([np.uint8(np.clip(i,0,255)) for i in image_with_noise])

# Plot our noisy image!
fig = plt.figure(figsize = (8,8))
ax1 = fig.add_subplot(111)
ax1.set_xticks([])
ax1.set_yticks([])

ax1.set_title('Noisy Image')
ax1.imshow(image_with_noise)

<matplotlib.image.AxesImage at 0x7ff0823d35f8>

In the context of face detection, the problem with an image like this is that - due to noise - we may miss some faces or get false detections.

In the next cell we apply the same trained OpenCV detector with the same settings as before, to see what sort of detections we get.

def detect_faces(image, sens=4):
    # Convert the RGB  image to grayscale
    gray_noise = cv2.cvtColor(image, cv2.COLOR_RGB2GRAY)

    # Extract the pre-trained face detector from an xml file
    face_cascade = cv2.CascadeClassifier('detector_architectures/haarcascade_frontalface_default.xml')

    # Detect the faces in image
    faces = face_cascade.detectMultiScale(gray_noise, sens, 6)

    # Print the number of faces detected in the image
    print('Number of faces detected:', len(faces))

    # Make a copy of the orginal image to draw face detections on
    image_with_detections = np.copy(image)

    # Get the bounding box for each detected face
    for (x,y,w,h) in faces:
        # Add a red bounding box to the detections image
        cv2.rectangle(image_with_detections, (x,y), (x+w,y+h), (255,0,0), 3)
    
    return image_with_detections

image_with_detections = detect_faces(image_with_noise)

# Display the image with the detections
fig = plt.figure(figsize = (8,8))
ax1 = fig.add_subplot(111)
ax1.set_xticks([])
ax1.set_yticks([])

ax1.set_title('Noisy Image with Face Detections')
ax1.imshow(image_with_detections)

Number of faces detected: 12

<matplotlib.image.AxesImage at 0x7ff081cf5c18>

With this added noise we now miss one of the faces!

(IMPLEMENTATION) De-noise this image for better face detection

Time to get your hands dirty: using OpenCV's built in color image de-noising functionality called fastNlMeansDenoisingColored - de-noise this image enough so that all the faces in the image are properly detected. Once you have cleaned the image in the next cell, use the cell that follows to run our trained face detector over the cleaned image to check out its detections.

You can find its official documentation here and a useful example here.

Note: you can keep all parameters except photo_render fixed as shown in the second link above. Play around with the value of this parameter - see how it affects the resulting cleaned image.

## Use OpenCV's built in color image de-noising function to clean up our noisy image!
denoised = cv2.fastNlMeansDenoisingColored(image_with_noise, None, 13, 13, 7, 21) # your final de-noised image (should be RGB)
# I could not get it up to 13 faces detected with any parameters (unlucky random noise)
# so I used blur as learned in lesson 3.24 to clear up the noise a bit more
denoised_image = cv2.GaussianBlur(denoised, (5,5), 0)

## Run the face detector on the de-noised image to improve your detections and display the result
denoised_faces = detect_faces(denoised_image)

# Display the image with the detections
fig = plt.figure(figsize = (8,8))
ax1 = fig.add_subplot(111)
ax1.set_xticks([])
ax1.set_yticks([])

ax1.set_title('Denoised Face Detections')
ax1.imshow(denoised_faces)

Number of faces detected: 13

<matplotlib.image.AxesImage at 0x7ff081c2a518>

---

Step 3: Blur an Image and Perform Edge Detection

Now that we have developed a simple pipeline for detecting faces using OpenCV - let's start playing around with a few fun things we can do with all those detected faces!

Importance of Blur in Edge Detection

Edge detection is a concept that pops up almost everywhere in computer vision applications, as edge-based features (as well as features built on top of edges) are often some of the best features for e.g., object detection and recognition problems.

Edge detection is a dimension reduction technique - by keeping only the edges of an image we get to throw away a lot of non-discriminating information. And typically the most useful kind of edge-detection is one that preserves only the important, global structures (ignoring local structures that aren't very discriminative). So removing local structures / retaining global structures is a crucial pre-processing step to performing edge detection in an image, and blurring can do just that.

Below is an animated gif showing the result of an edge-detected cat taken from Wikipedia, where the image is gradually blurred more and more prior to edge detection. When the animation begins you can't quite make out what it's a picture of, but as the animation evolves and local structures are removed via blurring the cat becomes visible in the edge-detected image.

Edge detection is a convolution performed on the image itself, and you can read about Canny edge detection on this OpenCV documentation page.

Canny edge detection

In the cell below we load in a test image, then apply Canny edge detection on it. The original image is shown on the left panel of the figure, while the edge-detected version of the image is shown on the right. Notice how the result looks very busy - there are too many little details preserved in the image before it is sent to the edge detector. When applied in computer vision applications, edge detection should preserve global structure; doing away with local structures that don't help describe what objects are in the image.

# Load in the image
image = cv2.imread('images/fawzia.jpg')

# Convert to RGB colorspace
image = cv2.cvtColor(image, cv2.COLOR_BGR2RGB)

# Convert to grayscale
gray = cv2.cvtColor(image, cv2.COLOR_RGB2GRAY)

# Perform Canny edge detection
edges = cv2.Canny(gray,100,200)

# Dilate the image to amplify edges
edges = cv2.dilate(edges, None)

# Plot the RGB and edge-detected image
fig = plt.figure(figsize = (15,15))
ax1 = fig.add_subplot(121)
ax1.set_xticks([])
ax1.set_yticks([])

ax1.set_title('Original Image')
ax1.imshow(image)

ax2 = fig.add_subplot(122)
ax2.set_xticks([])
ax2.set_yticks([])

ax2.set_title('Canny Edges')
ax2.imshow(edges, cmap='gray')

<matplotlib.image.AxesImage at 0x7f9e3420a9b0>

Without first blurring the image, and removing small, local structures, a lot of irrelevant edge content gets picked up and amplified by the detector (as shown in the right panel above).

(IMPLEMENTATION) Blur the image then perform edge detection

In the next cell, you will repeat this experiment - blurring the image first to remove these local structures, so that only the important boudnary details remain in the edge-detected image.

Blur the image by using OpenCV's filter2d functionality - which is discussed in this documentation page - and use an averaging kernel of width equal to 4.

### Blur the test image using OpenCV's filter2d functionality, 
# Use an averaging kernel, and a kernel width equal to 4
def blurring_kernel(gray_image, kernel_width):
    kernel = np.ones((kernel_width, kernel_width), np.float32) / kernel_width ** 2
    filtered_image = cv2.filter2D(gray_image, -1, kernel)
    return filtered_image

filtered = blurring_kernel(gray, 4)
# Perform Canny edge detection
edges = cv2.Canny(filtered, 100, 200)

# Dilate the image to amplify edges
edges = cv2.dilate(edges, None)

# Plot the RGB and edge-detected image
fig = plt.figure(figsize = (15,15))
ax1 = fig.add_subplot(121)
ax1.set_xticks([])
ax1.set_yticks([])

ax1.set_title('Original Image')
ax1.imshow(image)

ax2 = fig.add_subplot(122)
ax2.set_xticks([])
ax2.set_yticks([])

ax2.set_title('Canny Edges')
ax2.imshow(edges, cmap='gray')

<matplotlib.image.AxesImage at 0x7f9e3413a898>

---

Step 4: Automatically Hide the Identity of an Individual

If you film something like a documentary or reality TV, you must get permission from every individual shown on film before you can show their face, otherwise you need to blur it out - by blurring the face a lot (so much so that even the global structures are obscured)! This is also true for projects like Google's StreetView maps - an enormous collection of mapping images taken from a fleet of Google vehicles. Because it would be impossible for Google to get the permission of every single person accidentally captured in one of these images they blur out everyone's faces, the detected images must automatically blur the identity of detected people. Here's a few examples of folks caught in the camera of a Google street view vehicle.

Read in an image to perform identity detection

Let's try this out for ourselves. Use the face detection pipeline built above and what you know about using the filter2D to blur and image, and use these in tandem to hide the identity of the person in the following image - loaded in and printed in the next cell.

# Load in the image
image_gus = cv2.imread('images/gus.jpg')

# Convert the image to RGB colorspace
image_gus = cv2.cvtColor(image_gus, cv2.COLOR_BGR2RGB)

# Display the image
fig = plt.figure(figsize = (6,6))
ax1 = fig.add_subplot(111)
ax1.set_xticks([])
ax1.set_yticks([])
ax1.set_title('Original Image')
ax1.imshow(image_gus)

<matplotlib.image.AxesImage at 0x7f9e35d9e668>

(IMPLEMENTATION) Use blurring to hide the identity of an individual in an image

The idea here is to 1) automatically detect the face in this image, and then 2) blur it out! Make sure to adjust the parameters of the averaging blur filter to completely obscure this person's identity.

## Implement face detection
# Convert the RGB  image to grayscale
gray_gus = cv2.cvtColor(image_gus, cv2.COLOR_RGB2GRAY)

# Extract the pre-trained face detector from an xml file
face_cascade = cv2.CascadeClassifier('detector_architectures/haarcascade_frontalface_default.xml')

# Detect the faces in image
face_gus = face_cascade.detectMultiScale(gray_gus, 1.4, 4)

# Make a copy of the orginal image to draw face detections on
image_with_detections = np.copy(image_gus)

## Blur the bounding box around each detected face using an averaging filter and display the result
# Get the bounding box for each detected face
for (x,y,w,h) in face_gus:
    # Add a red bounding box to the detections image
    #cv2.rectangle(image_with_detections, (x,y), (x+w,y+h), (255,0,0), 5)
    blurred_face = blurring_kernel(image_gus[y:y+h, x:x+w], 80)
    image_with_detections[y:y+h, x:x+w] = blurred_face

# Display the image with the detections
fig = plt.figure(figsize = (6,6))
ax1 = fig.add_subplot(111)
ax1.set_xticks([])
ax1.set_yticks([])

ax1.imshow(image_with_detections)

<matplotlib.image.AxesImage at 0x7f9e35d26a90>

(Optional) Build identity protection into your laptop camera

In this optional task you can add identity protection to your laptop camera, using the previously completed code where you added face detection to your laptop camera - and the task above. You should be able to get reasonable results with little parameter tuning - like the one shown in the gif below.

As with the previous video task, to make this perfect would require significant effort - so don't strive for perfection here, strive for reasonable quality.

The next cell contains code a wrapper function called laptop_camera_identity_hider that - when called - will activate your laptop's camera. You need to place the relevant face detection and blurring code developed above in this function in order to blur faces entering your laptop camera's field of view.

Before adding anything to the function you can call it to get a hang of how it works - a small window will pop up showing you the live feed from your camera, you can press any key to close this window.

Note: Mac users may find that activating this function kills the kernel of their notebook every once in a while. If this happens to you, just restart your notebook's kernel, activate cell(s) containing any crucial import statements, and you'll be good to go!

### Insert face detection and blurring code into the wrapper below to create an identity protector on your laptop!
import cv2
import time

# Extract the pre-trained face detector from an xml file
face_cascade = cv2.CascadeClassifier('detector_architectures/haarcascade_frontalface_default.xml')

def laptop_camera_go():
    # Create instance of video capturer
    #cv2.namedWindow("face detection activated")
    vc = cv2.VideoCapture(0)

    # Try to get the first frame
    if vc.isOpened(): 
        rval, frame = vc.read()
    else:
        rval = False
    
    # Keep video stream open
    while rval:
        
        # Convert the RGB  image to grayscale
        gray_image = cv2.cvtColor(frame, cv2.COLOR_RGB2GRAY)

        # Detect the faces in image
        faces = face_cascade.detectMultiScale(gray_image, 1.4, 4)

        # Make a copy of the orginal image to draw face detections on
        image_with_detections = np.copy(frame)

        ## Blur the bounding box around each detected face using an averaging filter and display the result
        # Get the bounding box for each detected face
        for (x,y,w,h) in frame:
            # Add a red bounding box to the detections image
            #cv2.rectangle(image_with_detections, (x,y), (x+w,y+h), (255,0,0), 5)
            blurred_face = blurring_kernel(frame[y:y+h, x:x+w], 80)
            image_with_detections[y:y+h, x:x+w] = blurred_face
        
        # Plot image from camera with detections marked
        cv2.imshow("face detection activated", frame)
        
        # Exit functionality - press any key to exit laptop video
        key = cv2.waitKey(20)
        if key > 0: # Exit by pressing any key
            # Destroy windows
            cv2.destroyAllWindows()
            
            for i in range (1,5):
                cv2.waitKey(1)
            return
        
        # Read next frame
        time.sleep(0.05)             # control framerate for computation - default 20 frames per sec
        rval, frame = vc.read()    
        

# Run laptop identity hider
laptop_camera_go()

---

Step 5: Create a CNN to Recognize Facial Keypoints

OpenCV is often used in practice with other machine learning and deep learning libraries to produce interesting results. In this stage of the project you will create your own end-to-end pipeline - employing convolutional networks in keras along with OpenCV - to apply a "selfie" filter to streaming video and images.

You will start by creating and then training a convolutional network that can detect facial keypoints in a small dataset of cropped images of human faces. We then guide you towards OpenCV to expanding your detection algorithm to more general images. What are facial keypoints? Let's take a look at some examples.

Facial keypoints (also called facial landmarks) are the small blue-green dots shown on each of the faces in the image above - there are 15 keypoints marked in each image. They mark important areas of the face - the eyes, corners of the mouth, the nose, etc. Facial keypoints can be used in a variety of machine learning applications from face and emotion recognition to commercial applications like the image filters popularized by Snapchat.

Below we illustrate a filter that, using the results of this section, automatically places sunglasses on people in images (using the facial keypoints to place the glasses correctly on each face). Here, the facial keypoints have been colored lime green for visualization purposes.

Make a facial keypoint detector

But first things first: how can we make a facial keypoint detector? Well, at a high level, notice that facial keypoint detection is a regression problem. A single face corresponds to a set of 15 facial keypoints (a set of 15 corresponding $(x, y)$ coordinates, i.e., an output point). Because our input data are images, we can employ a convolutional neural network to recognize patterns in our images and learn how to identify these keypoint given sets of labeled data.

In order to train a regressor, we need a training set - a set of facial image / facial keypoint pairs to train on. For this we will be using this dataset from Kaggle. We've already downloaded this data and placed it in the data directory. Make sure that you have both the training and test data files. The training dataset contains several thousand $96 \times 96$ grayscale images of cropped human faces, along with each face's 15 corresponding facial keypoints (also called landmarks) that have been placed by hand, and recorded in $(x, y)$ coordinates. This wonderful resource also has a substantial testing set, which we will use in tinkering with our convolutional network.

To load in this data, run the Python cell below - notice we will load in both the training and testing sets.

The load_data function is in the included utils.py file.

from utils import *

# Load training set
X_train, y_train = load_data()
print("X_train.shape == {}".format(X_train.shape))
print("y_train.shape == {}; y_train.min == {:.3f}; y_train.max == {:.3f}".format(
    y_train.shape, y_train.min(), y_train.max()))

# Load testing set
X_test, _ = load_data(test=True)
print("X_test.shape == {}".format(X_test.shape))

X_train.shape == (2140, 96, 96, 1)
y_train.shape == (2140, 30); y_train.min == -0.920; y_train.max == 0.996
X_test.shape == (1783, 96, 96, 1)

The load_data function in utils.py originates from this excellent blog post, which you are strongly encouraged to read. Please take the time now to review this function. Note how the output values - that is, the coordinates of each set of facial landmarks - have been normalized to take on values in the range $[-1, 1]$, while the pixel values of each input point (a facial image) have been normalized to the range $[0,1]$.

Note: the original Kaggle dataset contains some images with several missing keypoints. For simplicity, the load_data function removes those images with missing labels from the dataset. As an optional extension, you are welcome to amend the load_data function to include the incomplete data points.

Visualize the Training Data

Execute the code cell below to visualize a subset of the training data.

import matplotlib.pyplot as plt
%matplotlib inline

fig = plt.figure(figsize=(20,20))
fig.subplots_adjust(left=0, right=1, bottom=0, top=1, hspace=0.05, wspace=0.05)
for i in range(9):
    ax = fig.add_subplot(3, 3, i + 1, xticks=[], yticks=[])
    plot_data(X_train[i], y_train[i], ax)

For each training image, there are two landmarks per eyebrow (four total), three per eye (six total), four for the mouth, and one for the tip of the nose.

Review the plot_data function in utils.py to understand how the 30-dimensional training labels in y_train are mapped to facial locations, as this function will prove useful for your pipeline.

(IMPLEMENTATION) Specify the CNN Architecture

In this section, you will specify a neural network for predicting the locations of facial keypoints. Use the code cell below to specify the architecture of your neural network. We have imported some layers that you may find useful for this task, but if you need to use more Keras layers, feel free to import them in the cell.

Your network should accept a $96 \times 96$ grayscale image as input, and it should output a vector with 30 entries, corresponding to the predicted (horizontal and vertical) locations of 15 facial keypoints. If you are not sure where to start, you can find some useful starting architectures in this blog, but you are not permitted to copy any of the architectures that you find online.

# Import deep learning resources from Keras
from keras.models import Sequential
from keras.layers import Convolution2D, MaxPooling2D, Dropout
from keras.layers import Flatten, Dense

## TODO: Specify a CNN architecture
# Your model should accept 96x96 pixel graysale images in
input_shape = (96, 96, 1)
# It should have a fully-connected output layer with 30 values (2 for each facial keypoint)
output_size = 30

model = Sequential()

#16
model.add(Convolution2D(filters=16, kernel_size=2, padding='same', activation='relu',input_shape=input_shape))
model.add(MaxPooling2D(pool_size=2))

#32
model.add(Convolution2D(filters=32, kernel_size=2, padding='same', activation='relu'))
model.add(MaxPooling2D(pool_size=2))

#64
model.add(Convolution2D(filters=64, kernel_size=2, padding='same', activation='relu'))
model.add(MaxPooling2D(pool_size=2))

#128
model.add(Convolution2D(filters=128, kernel_size=3, padding='same', activation='relu'))
model.add(MaxPooling2D(pool_size=2))

#256
model.add(Convolution2D(filters=256, kernel_size=4, padding='same', activation='relu'))
model.add(MaxPooling2D(pool_size=2))

model.add(Flatten())
model.add(Dense(512, activation='relu'))
model.add(Dropout(0.5))
model.add(Dense(output_size, activation=softmax))

# Summarize the model
model.summary()

_________________________________________________________________
Layer (type)                 Output Shape              Param #   
=================================================================
conv2d_6 (Conv2D)            (None, 96, 96, 16)        80        
_________________________________________________________________
max_pooling2d_6 (MaxPooling2 (None, 48, 48, 16)        0         
_________________________________________________________________
conv2d_7 (Conv2D)            (None, 48, 48, 32)        2080      
_________________________________________________________________
max_pooling2d_7 (MaxPooling2 (None, 24, 24, 32)        0         
_________________________________________________________________
conv2d_8 (Conv2D)            (None, 24, 24, 64)        8256      
_________________________________________________________________
max_pooling2d_8 (MaxPooling2 (None, 12, 12, 64)        0         
_________________________________________________________________
conv2d_9 (Conv2D)            (None, 12, 12, 128)       73856     
_________________________________________________________________
max_pooling2d_9 (MaxPooling2 (None, 6, 6, 128)         0         
_________________________________________________________________
conv2d_10 (Conv2D)           (None, 6, 6, 256)         524544    
_________________________________________________________________
max_pooling2d_10 (MaxPooling (None, 3, 3, 256)         0         
_________________________________________________________________
flatten_2 (Flatten)          (None, 2304)              0         
_________________________________________________________________
dense_3 (Dense)              (None, 512)               1180160   
_________________________________________________________________
dropout_2 (Dropout)          (None, 512)               0         
_________________________________________________________________
dense_4 (Dense)              (None, 30)                15390     
=================================================================
Total params: 1,804,366
Trainable params: 1,804,366
Non-trainable params: 0
_________________________________________________________________

---

Step 6: Compile and Train the Model

After specifying your architecture, you'll need to compile and train the model to detect facial keypoints'

(IMPLEMENTATION) Compile and Train the Model

Use the compile method to configure the learning process. Experiment with your choice of optimizer; you may have some ideas about which will work best (SGD vs. RMSprop, etc), but take the time to empirically verify your theories.

Use the fit method to train the model. Break off a validation set by setting validation_split=0.2. Save the returned History object in the history variable.

Experiment with your model to minimize the validation loss (measured as mean squared error). A very good model will achieve about 0.0015 loss (though it's possible to do even better). When you have finished training, save your model as an HDF5 file with file path my_model.h5.

from keras.optimizers import SGD, RMSprop, Adagrad, Adadelta, Adam, Adamax, Nadam
from keras.callbacks import ModelCheckpoint
from keras.models import load_model
from datetime import datetime

optimizers_names = [SGD, RMSprop, Adagrad, Adadelta, Adam, Adamax, Nadam]
# learning_rates = [0.00001, 0.0001, 0.001]

optimizers = dict([(t.__name__, t) for t in optimizers_names])

nr_of_epochs = 100
size_of_batch = 32

def model_load_weights(the_model, filename):
    import os.path
    exists = os.path.isfile(filename)
    if exists:
        print("Loading best weights...")
        the_model.load_weights(filename)
    else:
        print("Did not load old weights...")
        pass
    
model_histories = {}

for optimizer in optimizers:
    print("\n\n", datetime.now())
    print("Using optimizer: ", optimizer)
    model_filename = 'models/my_model_' + optimizer + '.h5'
    model_weights_filename = 'weights/modelweights_' + optimizer + '.h5'
    
    model_load_weights(model, model_weights_filename)
    ## Compile the model
    model.compile(optimizer=optimizers[optimizer](), loss='mean_squared_error', metrics=['accuracy'])
    
    checkpointer = ModelCheckpoint(filepath=model_weights_filename, verbose=1, save_best_only=True)
    
    ## Train the model
    model_history = model.fit(X_train, y_train, 
                     batch_size=size_of_batch, epochs=nr_of_epochs, 
                     validation_split=0.2,
                     verbose=1,
                     callbacks=[checkpointer])
    
    model_histories[optimizer] = model_history
    
    model.save(model_filename)

 2017-12-11 23:23:08.175114
Using optimizer:  Nadam
Did not load old weights...
Train on 1712 samples, validate on 428 samples
Epoch 1/100
1696/1712 [============================>.] - ETA: 0s - loss: 0.0016 - acc: 0.7353Epoch 00001: val_loss improved from inf to 0.00166, saving model to weights/modelweights_Nadam.h5
1712/1712 [==============================] - 3s 2ms/step - loss: 0.0016 - acc: 0.7342 - val_loss: 0.0017 - val_acc: 0.7430
Epoch 2/100
1696/1712 [============================>.] - ETA: 0s - loss: 0.0016 - acc: 0.7465Epoch 00002: val_loss improved from 0.00166 to 0.00163, saving model to weights/modelweights_Nadam.h5
1712/1712 [==============================] - 2s 1ms/step - loss: 0.0016 - acc: 0.7465 - val_loss: 0.0016 - val_acc: 0.7290
Epoch 3/100
1696/1712 [============================>.] - ETA: 0s - loss: 0.0015 - acc: 0.7429Epoch 00003: val_loss did not improve
1712/1712 [==============================] - 2s 1ms/step - loss: 0.0015 - acc: 0.7407 - val_loss: 0.0017 - val_acc: 0.7079
Epoch 4/100
1696/1712 [============================>.] - ETA: 0s - loss: 0.0015 - acc: 0.7553Epoch 00004: val_loss improved from 0.00163 to 0.00157, saving model to weights/modelweights_Nadam.h5
1712/1712 [==============================] - 2s 1ms/step - loss: 0.0015 - acc: 0.7541 - val_loss: 0.0016 - val_acc: 0.7313
Epoch 5/100
1696/1712 [============================>.] - ETA: 0s - loss: 0.0015 - acc: 0.7500Epoch 00005: val_loss did not improve
1712/1712 [==============================] - 2s 1ms/step - loss: 0.0015 - acc: 0.7494 - val_loss: 0.0018 - val_acc: 0.7173
Epoch 6/100
1696/1712 [============================>.] - ETA: 0s - loss: 0.0015 - acc: 0.7453Epoch 00006: val_loss did not improve
1712/1712 [==============================] - 2s 1ms/step - loss: 0.0015 - acc: 0.7447 - val_loss: 0.0016 - val_acc: 0.7173
Epoch 7/100
1696/1712 [============================>.] - ETA: 0s - loss: 0.0014 - acc: 0.7547Epoch 00007: val_loss did not improve
1712/1712 [==============================] - 2s 1ms/step - loss: 0.0014 - acc: 0.7541 - val_loss: 0.0017 - val_acc: 0.7430
Epoch 8/100
1696/1712 [============================>.] - ETA: 0s - loss: 0.0014 - acc: 0.7459Epoch 00008: val_loss improved from 0.00157 to 0.00156, saving model to weights/modelweights_Nadam.h5
1712/1712 [==============================] - 2s 1ms/step - loss: 0.0014 - acc: 0.7471 - val_loss: 0.0016 - val_acc: 0.7266
Epoch 9/100
1696/1712 [============================>.] - ETA: 0s - loss: 0.0014 - acc: 0.7612Epoch 00009: val_loss improved from 0.00156 to 0.00155, saving model to weights/modelweights_Nadam.h5
1712/1712 [==============================] - 2s 1ms/step - loss: 0.0014 - acc: 0.7617 - val_loss: 0.0015 - val_acc: 0.7220
Epoch 10/100
1696/1712 [============================>.] - ETA: 0s - loss: 0.0015 - acc: 0.7524Epoch 00010: val_loss did not improve
1712/1712 [==============================] - 2s 1ms/step - loss: 0.0015 - acc: 0.7512 - val_loss: 0.0016 - val_acc: 0.7243
Epoch 11/100
1696/1712 [============================>.] - ETA: 0s - loss: 0.0013 - acc: 0.7624Epoch 00011: val_loss did not improve
1712/1712 [==============================] - 2s 1ms/step - loss: 0.0013 - acc: 0.7646 - val_loss: 0.0018 - val_acc: 0.7290
Epoch 12/100
1696/1712 [============================>.] - ETA: 0s - loss: 0.0014 - acc: 0.7535Epoch 00012: val_loss improved from 0.00155 to 0.00151, saving model to weights/modelweights_Nadam.h5
1712/1712 [==============================] - 2s 1ms/step - loss: 0.0014 - acc: 0.7529 - val_loss: 0.0015 - val_acc: 0.7523
Epoch 13/100
1696/1712 [============================>.] - ETA: 0s - loss: 0.0014 - acc: 0.7612Epoch 00013: val_loss did not improve
1712/1712 [==============================] - 2s 1ms/step - loss: 0.0014 - acc: 0.7623 - val_loss: 0.0016 - val_acc: 0.7523
Epoch 14/100
1696/1712 [============================>.] - ETA: 0s - loss: 0.0013 - acc: 0.7677Epoch 00014: val_loss did not improve
1712/1712 [==============================] - 2s 1ms/step - loss: 0.0013 - acc: 0.7675 - val_loss: 0.0015 - val_acc: 0.7383
Epoch 15/100
1696/1712 [============================>.] - ETA: 0s - loss: 0.0013 - acc: 0.7588Epoch 00015: val_loss did not improve
1712/1712 [==============================] - 2s 1ms/step - loss: 0.0013 - acc: 0.7588 - val_loss: 0.0015 - val_acc: 0.7407
Epoch 16/100
1696/1712 [============================>.] - ETA: 0s - loss: 0.0013 - acc: 0.7529Epoch 00016: val_loss improved from 0.00151 to 0.00146, saving model to weights/modelweights_Nadam.h5
1712/1712 [==============================] - 2s 1ms/step - loss: 0.0013 - acc: 0.7523 - val_loss: 0.0015 - val_acc: 0.7313
Epoch 17/100
1696/1712 [============================>.] - ETA: 0s - loss: 0.0013 - acc: 0.7642Epoch 00017: val_loss did not improve
1712/1712 [==============================] - 2s 1ms/step - loss: 0.0013 - acc: 0.7634 - val_loss: 0.0015 - val_acc: 0.7196
Epoch 18/100
1696/1712 [============================>.] - ETA: 0s - loss: 0.0013 - acc: 0.7677Epoch 00018: val_loss did not improve
1712/1712 [==============================] - 2s 1ms/step - loss: 0.0013 - acc: 0.7681 - val_loss: 0.0016 - val_acc: 0.7407
Epoch 19/100
1696/1712 [============================>.] - ETA: 0s - loss: 0.0012 - acc: 0.7712Epoch 00019: val_loss did not improve
1712/1712 [==============================] - 2s 1ms/step - loss: 0.0012 - acc: 0.7699 - val_loss: 0.0015 - val_acc: 0.7407
Epoch 20/100
1696/1712 [============================>.] - ETA: 0s - loss: 0.0013 - acc: 0.7706Epoch 00020: val_loss did not improve
1712/1712 [==============================] - 2s 1ms/step - loss: 0.0013 - acc: 0.7710 - val_loss: 0.0015 - val_acc: 0.7313
Epoch 21/100
1696/1712 [============================>.] - ETA: 0s - loss: 0.0012 - acc: 0.7689Epoch 00021: val_loss did not improve
1712/1712 [==============================] - 2s 1ms/step - loss: 0.0012 - acc: 0.7699 - val_loss: 0.0016 - val_acc: 0.7593
Epoch 22/100
1696/1712 [============================>.] - ETA: 0s - loss: 0.0012 - acc: 0.7689Epoch 00022: val_loss did not improve
1712/1712 [==============================] - 2s 1ms/step - loss: 0.0012 - acc: 0.7664 - val_loss: 0.0015 - val_acc: 0.7593
Epoch 23/100
1696/1712 [============================>.] - ETA: 0s - loss: 0.0012 - acc: 0.7642Epoch 00023: val_loss improved from 0.00146 to 0.00146, saving model to weights/modelweights_Nadam.h5
1712/1712 [==============================] - 2s 1ms/step - loss: 0.0012 - acc: 0.7634 - val_loss: 0.0015 - val_acc: 0.7453
Epoch 24/100
1696/1712 [============================>.] - ETA: 0s - loss: 0.0012 - acc: 0.7795Epoch 00024: val_loss improved from 0.00146 to 0.00145, saving model to weights/modelweights_Nadam.h5
1712/1712 [==============================] - 2s 1ms/step - loss: 0.0012 - acc: 0.7786 - val_loss: 0.0015 - val_acc: 0.7477
Epoch 25/100
1696/1712 [============================>.] - ETA: 0s - loss: 0.0012 - acc: 0.7683Epoch 00025: val_loss did not improve
1712/1712 [==============================] - 2s 1ms/step - loss: 0.0012 - acc: 0.7687 - val_loss: 0.0015 - val_acc: 0.7687
Epoch 26/100
1696/1712 [============================>.] - ETA: 0s - loss: 0.0012 - acc: 0.7695Epoch 00026: val_loss did not improve
1712/1712 [==============================] - 2s 1ms/step - loss: 0.0012 - acc: 0.7687 - val_loss: 0.0015 - val_acc: 0.7477
Epoch 27/100
1696/1712 [============================>.] - ETA: 0s - loss: 0.0013 - acc: 0.7600Epoch 00027: val_loss did not improve
1712/1712 [==============================] - 2s 1ms/step - loss: 0.0013 - acc: 0.7599 - val_loss: 0.0017 - val_acc: 0.7220
Epoch 28/100
1696/1712 [============================>.] - ETA: 0s - loss: 0.0012 - acc: 0.7748Epoch 00028: val_loss improved from 0.00145 to 0.00139, saving model to weights/modelweights_Nadam.h5
1712/1712 [==============================] - 2s 1ms/step - loss: 0.0012 - acc: 0.7769 - val_loss: 0.0014 - val_acc: 0.7500
Epoch 29/100
1696/1712 [============================>.] - ETA: 0s - loss: 0.0012 - acc: 0.7736Epoch 00029: val_loss did not improve
1712/1712 [==============================] - 2s 1ms/step - loss: 0.0012 - acc: 0.7722 - val_loss: 0.0014 - val_acc: 0.7547
Epoch 30/100
1696/1712 [============================>.] - ETA: 0s - loss: 0.0012 - acc: 0.7765Epoch 00030: val_loss did not improve
1712/1712 [==============================] - 2s 1ms/step - loss: 0.0012 - acc: 0.7763 - val_loss: 0.0015 - val_acc: 0.7383
Epoch 31/100
1696/1712 [============================>.] - ETA: 0s - loss: 0.0012 - acc: 0.7706Epoch 00031: val_loss did not improve
1712/1712 [==============================] - 2s 1ms/step - loss: 0.0012 - acc: 0.7704 - val_loss: 0.0016 - val_acc: 0.7453
Epoch 32/100
1696/1712 [============================>.] - ETA: 0s - loss: 0.0012 - acc: 0.7700Epoch 00032: val_loss did not improve
1712/1712 [==============================] - 2s 1ms/step - loss: 0.0012 - acc: 0.7699 - val_loss: 0.0014 - val_acc: 0.7523
Epoch 33/100
1696/1712 [============================>.] - ETA: 0s - loss: 0.0012 - acc: 0.7824Epoch 00033: val_loss did not improve
1712/1712 [==============================] - 2s 1ms/step - loss: 0.0012 - acc: 0.7810 - val_loss: 0.0014 - val_acc: 0.7617
Epoch 34/100
1696/1712 [============================>.] - ETA: 0s - loss: 0.0012 - acc: 0.7724Epoch 00034: val_loss did not improve
1712/1712 [==============================] - 2s 1ms/step - loss: 0.0012 - acc: 0.7728 - val_loss: 0.0014 - val_acc: 0.7360
Epoch 35/100
1696/1712 [============================>.] - ETA: 0s - loss: 0.0012 - acc: 0.7830Epoch 00035: val_loss did not improve
1712/1712 [==============================] - 2s 1ms/step - loss: 0.0012 - acc: 0.7839 - val_loss: 0.0014 - val_acc: 0.7617
Epoch 36/100
1696/1712 [============================>.] - ETA: 0s - loss: 0.0011 - acc: 0.7818Epoch 00036: val_loss did not improve
1712/1712 [==============================] - 2s 1ms/step - loss: 0.0011 - acc: 0.7821 - val_loss: 0.0015 - val_acc: 0.7407
Epoch 37/100
1696/1712 [============================>.] - ETA: 0s - loss: 0.0012 - acc: 0.7812Epoch 00037: val_loss did not improve
1712/1712 [==============================] - 2s 1ms/step - loss: 0.0012 - acc: 0.7821 - val_loss: 0.0014 - val_acc: 0.7617
Epoch 38/100
1696/1712 [============================>.] - ETA: 0s - loss: 0.0011 - acc: 0.7724Epoch 00038: val_loss did not improve
1712/1712 [==============================] - 2s 1ms/step - loss: 0.0011 - acc: 0.7722 - val_loss: 0.0014 - val_acc: 0.7453
Epoch 39/100
1696/1712 [============================>.] - ETA: 0s - loss: 0.0011 - acc: 0.7883Epoch 00039: val_loss did not improve
1712/1712 [==============================] - 2s 1ms/step - loss: 0.0011 - acc: 0.7886 - val_loss: 0.0014 - val_acc: 0.7640
Epoch 40/100
1696/1712 [============================>.] - ETA: 0s - loss: 0.0011 - acc: 0.7883Epoch 00040: val_loss did not improve
1712/1712 [==============================] - 2s 1ms/step - loss: 0.0011 - acc: 0.7880 - val_loss: 0.0014 - val_acc: 0.7477
Epoch 41/100
1696/1712 [============================>.] - ETA: 0s - loss: 0.0011 - acc: 0.7818Epoch 00041: val_loss did not improve
1712/1712 [==============================] - 2s 1ms/step - loss: 0.0012 - acc: 0.7815 - val_loss: 0.0014 - val_acc: 0.7687
Epoch 42/100
1696/1712 [============================>.] - ETA: 0s - loss: 0.0011 - acc: 0.7742Epoch 00042: val_loss did not improve
1712/1712 [==============================] - 2s 1ms/step - loss: 0.0011 - acc: 0.7722 - val_loss: 0.0015 - val_acc: 0.7500
Epoch 43/100
1696/1712 [============================>.] - ETA: 0s - loss: 0.0011 - acc: 0.7883Epoch 00043: val_loss did not improve
1712/1712 [==============================] - 2s 1ms/step - loss: 0.0011 - acc: 0.7886 - val_loss: 0.0014 - val_acc: 0.7757
Epoch 44/100
1696/1712 [============================>.] - ETA: 0s - loss: 0.0011 - acc: 0.7848Epoch 00044: val_loss improved from 0.00139 to 0.00137, saving model to weights/modelweights_Nadam.h5
1712/1712 [==============================] - 2s 1ms/step - loss: 0.0011 - acc: 0.7868 - val_loss: 0.0014 - val_acc: 0.7593
Epoch 45/100
1696/1712 [============================>.] - ETA: 0s - loss: 0.0011 - acc: 0.7836Epoch 00045: val_loss did not improve
1712/1712 [==============================] - 2s 1ms/step - loss: 0.0011 - acc: 0.7821 - val_loss: 0.0014 - val_acc: 0.7383
Epoch 46/100
1696/1712 [============================>.] - ETA: 0s - loss: 0.0011 - acc: 0.7659Epoch 00046: val_loss did not improve
1712/1712 [==============================] - 2s 1ms/step - loss: 0.0011 - acc: 0.7664 - val_loss: 0.0015 - val_acc: 0.7593
Epoch 47/100
1696/1712 [============================>.] - ETA: 0s - loss: 0.0011 - acc: 0.7671Epoch 00047: val_loss did not improve
1712/1712 [==============================] - 2s 1ms/step - loss: 0.0011 - acc: 0.7658 - val_loss: 0.0015 - val_acc: 0.7477
Epoch 48/100
1696/1712 [============================>.] - ETA: 0s - loss: 0.0011 - acc: 0.7742Epoch 00048: val_loss did not improve
1712/1712 [==============================] - 2s 1ms/step - loss: 0.0011 - acc: 0.7739 - val_loss: 0.0014 - val_acc: 0.7664
Epoch 49/100
1696/1712 [============================>.] - ETA: 0s - loss: 0.0011 - acc: 0.7783Epoch 00049: val_loss did not improve
1712/1712 [==============================] - 2s 1ms/step - loss: 0.0011 - acc: 0.7763 - val_loss: 0.0015 - val_acc: 0.7710
Epoch 50/100
1696/1712 [============================>.] - ETA: 0s - loss: 0.0011 - acc: 0.7777Epoch 00050: val_loss did not improve
1712/1712 [==============================] - 2s 1ms/step - loss: 0.0011 - acc: 0.7798 - val_loss: 0.0014 - val_acc: 0.7593
Epoch 51/100
1696/1712 [============================>.] - ETA: 0s - loss: 0.0011 - acc: 0.7712Epoch 00051: val_loss did not improve
1712/1712 [==============================] - 2s 1ms/step - loss: 0.0011 - acc: 0.7734 - val_loss: 0.0014 - val_acc: 0.7664
Epoch 52/100
1696/1712 [============================>.] - ETA: 0s - loss: 0.0011 - acc: 0.7801Epoch 00052: val_loss did not improve
1712/1712 [==============================] - 2s 1ms/step - loss: 0.0011 - acc: 0.7792 - val_loss: 0.0014 - val_acc: 0.7664
Epoch 53/100
1696/1712 [============================>.] - ETA: 0s - loss: 0.0011 - acc: 0.7736Epoch 00053: val_loss did not improve
1712/1712 [==============================] - 2s 1ms/step - loss: 0.0011 - acc: 0.7734 - val_loss: 0.0014 - val_acc: 0.7523
Epoch 54/100
1696/1712 [============================>.] - ETA: 0s - loss: 0.0011 - acc: 0.7824Epoch 00054: val_loss did not improve
1712/1712 [==============================] - 2s 1ms/step - loss: 0.0011 - acc: 0.7815 - val_loss: 0.0015 - val_acc: 0.7430
Epoch 55/100
1696/1712 [============================>.] - ETA: 0s - loss: 0.0011 - acc: 0.7777Epoch 00055: val_loss did not improve
1712/1712 [==============================] - 2s 1ms/step - loss: 0.0011 - acc: 0.7780 - val_loss: 0.0015 - val_acc: 0.7430
Epoch 56/100
1696/1712 [============================>.] - ETA: 0s - loss: 0.0011 - acc: 0.7730Epoch 00056: val_loss did not improve
1712/1712 [==============================] - 2s 1ms/step - loss: 0.0011 - acc: 0.7739 - val_loss: 0.0015 - val_acc: 0.7383
Epoch 57/100
1696/1712 [============================>.] - ETA: 0s - loss: 0.0011 - acc: 0.7812Epoch 00057: val_loss did not improve
1712/1712 [==============================] - 2s 1ms/step - loss: 0.0011 - acc: 0.7821 - val_loss: 0.0014 - val_acc: 0.7757
Epoch 58/100
1696/1712 [============================>.] - ETA: 0s - loss: 0.0011 - acc: 0.7866Epoch 00058: val_loss did not improve
1712/1712 [==============================] - 2s 1ms/step - loss: 0.0011 - acc: 0.7862 - val_loss: 0.0014 - val_acc: 0.7687
Epoch 59/100
1696/1712 [============================>.] - ETA: 0s - loss: 0.0011 - acc: 0.7748Epoch 00059: val_loss did not improve
1712/1712 [==============================] - 2s 1ms/step - loss: 0.0011 - acc: 0.7751 - val_loss: 0.0014 - val_acc: 0.7453
Epoch 60/100
1696/1712 [============================>.] - ETA: 0s - loss: 0.0011 - acc: 0.7848Epoch 00060: val_loss did not improve
1712/1712 [==============================] - 2s 1ms/step - loss: 0.0011 - acc: 0.7833 - val_loss: 0.0015 - val_acc: 0.7664
Epoch 61/100
1696/1712 [============================>.] - ETA: 0s - loss: 0.0011 - acc: 0.7748Epoch 00061: val_loss did not improve
1712/1712 [==============================] - 2s 1ms/step - loss: 0.0011 - acc: 0.7745 - val_loss: 0.0014 - val_acc: 0.7710
Epoch 62/100
1696/1712 [============================>.] - ETA: 0s - loss: 0.0011 - acc: 0.7936Epoch 00062: val_loss did not improve
1712/1712 [==============================] - 2s 1ms/step - loss: 0.0011 - acc: 0.7938 - val_loss: 0.0014 - val_acc: 0.7570
Epoch 63/100
1696/1712 [============================>.] - ETA: 0s - loss: 0.0011 - acc: 0.7783Epoch 00063: val_loss did not improve
1712/1712 [==============================] - 2s 1ms/step - loss: 0.0011 - acc: 0.7798 - val_loss: 0.0014 - val_acc: 0.7477
Epoch 64/100
1696/1712 [============================>.] - ETA: 0s - loss: 0.0011 - acc: 0.7860Epoch 00064: val_loss did not improve
1712/1712 [==============================] - 2s 1ms/step - loss: 0.0011 - acc: 0.7874 - val_loss: 0.0015 - val_acc: 0.7617
Epoch 65/100
1696/1712 [============================>.] - ETA: 0s - loss: 0.0011 - acc: 0.7895Epoch 00065: val_loss did not improve
1712/1712 [==============================] - 2s 1ms/step - loss: 0.0011 - acc: 0.7903 - val_loss: 0.0014 - val_acc: 0.7640
Epoch 66/100
1696/1712 [============================>.] - ETA: 0s - loss: 0.0011 - acc: 0.7830Epoch 00066: val_loss did not improve
1712/1712 [==============================] - 2s 1ms/step - loss: 0.0011 - acc: 0.7821 - val_loss: 0.0014 - val_acc: 0.7430
Epoch 67/100
1696/1712 [============================>.] - ETA: 0s - loss: 0.0011 - acc: 0.7954Epoch 00067: val_loss improved from 0.00137 to 0.00137, saving model to weights/modelweights_Nadam.h5
1712/1712 [==============================] - 2s 1ms/step - loss: 0.0011 - acc: 0.7961 - val_loss: 0.0014 - val_acc: 0.7547
Epoch 68/100
1696/1712 [============================>.] - ETA: 0s - loss: 0.0011 - acc: 0.8048Epoch 00068: val_loss did not improve
1712/1712 [==============================] - 2s 1ms/step - loss: 0.0011 - acc: 0.8049 - val_loss: 0.0014 - val_acc: 0.7640
Epoch 69/100
1696/1712 [============================>.] - ETA: 0s - loss: 0.0011 - acc: 0.7854Epoch 00069: val_loss did not improve
1712/1712 [==============================] - 2s 1ms/step - loss: 0.0011 - acc: 0.7856 - val_loss: 0.0014 - val_acc: 0.7523
Epoch 70/100
1696/1712 [============================>.] - ETA: 0s - loss: 0.0011 - acc: 0.7866Epoch 00070: val_loss did not improve
1712/1712 [==============================] - 2s 1ms/step - loss: 0.0011 - acc: 0.7868 - val_loss: 0.0014 - val_acc: 0.7640
Epoch 71/100
1696/1712 [============================>.] - ETA: 0s - loss: 0.0011 - acc: 0.7818Epoch 00071: val_loss did not improve
1712/1712 [==============================] - 2s 1ms/step - loss: 0.0011 - acc: 0.7827 - val_loss: 0.0015 - val_acc: 0.7710
Epoch 72/100
1696/1712 [============================>.] - ETA: 0s - loss: 0.0010 - acc: 0.7907Epoch 00072: val_loss improved from 0.00137 to 0.00136, saving model to weights/modelweights_Nadam.h5
1712/1712 [==============================] - 2s 1ms/step - loss: 0.0010 - acc: 0.7915 - val_loss: 0.0014 - val_acc: 0.7780
Epoch 73/100
1696/1712 [============================>.] - ETA: 0s - loss: 0.0010 - acc: 0.7883Epoch 00073: val_loss did not improve
1712/1712 [==============================] - 2s 1ms/step - loss: 0.0010 - acc: 0.7891 - val_loss: 0.0014 - val_acc: 0.7593
Epoch 74/100
1696/1712 [============================>.] - ETA: 0s - loss: 0.0010 - acc: 0.7812Epoch 00074: val_loss did not improve
1712/1712 [==============================] - 2s 1ms/step - loss: 0.0010 - acc: 0.7821 - val_loss: 0.0014 - val_acc: 0.7780
Epoch 75/100
1696/1712 [============================>.] - ETA: 0s - loss: 0.0010 - acc: 0.7913Epoch 00075: val_loss did not improve
1712/1712 [==============================] - 2s 1ms/step - loss: 0.0010 - acc: 0.7909 - val_loss: 0.0014 - val_acc: 0.7640
Epoch 76/100
1696/1712 [============================>.] - ETA: 0s - loss: 0.0010 - acc: 0.7807Epoch 00076: val_loss did not improve
1712/1712 [==============================] - 2s 1ms/step - loss: 0.0010 - acc: 0.7798 - val_loss: 0.0014 - val_acc: 0.7430
Epoch 77/100
1696/1712 [============================>.] - ETA: 0s - loss: 0.0010 - acc: 0.7818Epoch 00077: val_loss did not improve
1712/1712 [==============================] - 2s 1ms/step - loss: 0.0010 - acc: 0.7821 - val_loss: 0.0014 - val_acc: 0.7477
Epoch 78/100
1696/1712 [============================>.] - ETA: 0s - loss: 0.0010 - acc: 0.7972Epoch 00078: val_loss did not improve
1712/1712 [==============================] - 2s 1ms/step - loss: 0.0010 - acc: 0.7961 - val_loss: 0.0014 - val_acc: 0.7710
Epoch 79/100
1696/1712 [============================>.] - ETA: 0s - loss: 0.0010 - acc: 0.7836Epoch 00079: val_loss did not improve
1712/1712 [==============================] - 2s 1ms/step - loss: 0.0010 - acc: 0.7845 - val_loss: 0.0014 - val_acc: 0.7547
Epoch 80/100
1696/1712 [============================>.] - ETA: 0s - loss: 0.0010 - acc: 0.7907Epoch 00080: val_loss did not improve
1712/1712 [==============================] - 2s 1ms/step - loss: 0.0010 - acc: 0.7909 - val_loss: 0.0014 - val_acc: 0.7547
Epoch 81/100
1696/1712 [============================>.] - ETA: 0s - loss: 0.0010 - acc: 0.7942Epoch 00081: val_loss did not improve
1712/1712 [==============================] - 2s 1ms/step - loss: 0.0010 - acc: 0.7932 - val_loss: 0.0014 - val_acc: 0.7570
Epoch 82/100
1696/1712 [============================>.] - ETA: 0s - loss: 0.0010 - acc: 0.7960Epoch 00082: val_loss did not improve
1712/1712 [==============================] - 2s 1ms/step - loss: 0.0010 - acc: 0.7967 - val_loss: 0.0014 - val_acc: 0.7593
Epoch 83/100
1696/1712 [============================>.] - ETA: 0s - loss: 0.0010 - acc: 0.7919Epoch 00083: val_loss did not improve
1712/1712 [==============================] - 2s 1ms/step - loss: 0.0010 - acc: 0.7909 - val_loss: 0.0014 - val_acc: 0.7617
Epoch 84/100
1696/1712 [============================>.] - ETA: 0s - loss: 0.0010 - acc: 0.7948Epoch 00084: val_loss did not improve
1712/1712 [==============================] - 2s 1ms/step - loss: 0.0010 - acc: 0.7938 - val_loss: 0.0014 - val_acc: 0.7593
Epoch 85/100
1696/1712 [============================>.] - ETA: 0s - loss: 0.0010 - acc: 0.7978Epoch 00085: val_loss did not improve
1712/1712 [==============================] - 2s 1ms/step - loss: 0.0010 - acc: 0.7973 - val_loss: 0.0014 - val_acc: 0.7547
Epoch 86/100
1696/1712 [============================>.] - ETA: 0s - loss: 0.0010 - acc: 0.8019Epoch 00086: val_loss did not improve
1712/1712 [==============================] - 2s 1ms/step - loss: 0.0010 - acc: 0.8020 - val_loss: 0.0014 - val_acc: 0.7523
Epoch 87/100
1696/1712 [============================>.] - ETA: 0s - loss: 0.0010 - acc: 0.7877Epoch 00087: val_loss did not improve
1712/1712 [==============================] - 2s 1ms/step - loss: 0.0010 - acc: 0.7874 - val_loss: 0.0014 - val_acc: 0.7710
Epoch 88/100
1696/1712 [============================>.] - ETA: 0s - loss: 0.0010 - acc: 0.7913Epoch 00088: val_loss did not improve
1712/1712 [==============================] - 2s 1ms/step - loss: 0.0010 - acc: 0.7926 - val_loss: 0.0014 - val_acc: 0.7710
Epoch 89/100
1696/1712 [============================>.] - ETA: 0s - loss: 0.0010 - acc: 0.7966Epoch 00089: val_loss did not improve
1712/1712 [==============================] - 2s 1ms/step - loss: 0.0010 - acc: 0.7967 - val_loss: 0.0015 - val_acc: 0.7664
Epoch 90/100
1696/1712 [============================>.] - ETA: 0s - loss: 0.0011 - acc: 0.7842Epoch 00090: val_loss improved from 0.00136 to 0.00136, saving model to weights/modelweights_Nadam.h5
1712/1712 [==============================] - 2s 1ms/step - loss: 0.0011 - acc: 0.7833 - val_loss: 0.0014 - val_acc: 0.7500
Epoch 91/100
1696/1712 [============================>.] - ETA: 0s - loss: 0.0010 - acc: 0.7877Epoch 00091: val_loss did not improve
1712/1712 [==============================] - 2s 1ms/step - loss: 0.0010 - acc: 0.7886 - val_loss: 0.0014 - val_acc: 0.7593
Epoch 92/100
1696/1712 [============================>.] - ETA: 0s - loss: 0.0010 - acc: 0.7848Epoch 00092: val_loss did not improve
1712/1712 [==============================] - 2s 1ms/step - loss: 0.0010 - acc: 0.7862 - val_loss: 0.0014 - val_acc: 0.7710
Epoch 93/100
1696/1712 [============================>.] - ETA: 0s - loss: 0.0010 - acc: 0.7812Epoch 00093: val_loss did not improve
1712/1712 [==============================] - 2s 1ms/step - loss: 0.0010 - acc: 0.7804 - val_loss: 0.0014 - val_acc: 0.7617
Epoch 94/100
1696/1712 [============================>.] - ETA: 0s - loss: 0.0011 - acc: 0.7824Epoch 00094: val_loss did not improve
1712/1712 [==============================] - 2s 1ms/step - loss: 0.0011 - acc: 0.7821 - val_loss: 0.0014 - val_acc: 0.7710
Epoch 95/100
1696/1712 [============================>.] - ETA: 0s - loss: 0.0010 - acc: 0.7901Epoch 00095: val_loss did not improve
1712/1712 [==============================] - 2s 1ms/step - loss: 0.0010 - acc: 0.7909 - val_loss: 0.0014 - val_acc: 0.7453
Epoch 96/100
1696/1712 [============================>.] - ETA: 0s - loss: 0.0010 - acc: 0.8007Epoch 00096: val_loss did not improve
1712/1712 [==============================] - 2s 1ms/step - loss: 0.0010 - acc: 0.7996 - val_loss: 0.0015 - val_acc: 0.7547
Epoch 97/100
1696/1712 [============================>.] - ETA: 0s - loss: 0.0010 - acc: 0.7854Epoch 00097: val_loss did not improve
1712/1712 [==============================] - 2s 1ms/step - loss: 0.0010 - acc: 0.7868 - val_loss: 0.0014 - val_acc: 0.7547
Epoch 98/100
1696/1712 [============================>.] - ETA: 0s - loss: 0.0010 - acc: 0.7877Epoch 00098: val_loss did not improve
1712/1712 [==============================] - 2s 1ms/step - loss: 0.0010 - acc: 0.7862 - val_loss: 0.0014 - val_acc: 0.7710
Epoch 99/100
1696/1712 [============================>.] - ETA: 0s - loss: 0.0010 - acc: 0.7818Epoch 00099: val_loss did not improve
1712/1712 [==============================] - 2s 1ms/step - loss: 0.0010 - acc: 0.7827 - val_loss: 0.0014 - val_acc: 0.7360
Epoch 100/100
1696/1712 [============================>.] - ETA: 0s - loss: 0.0010 - acc: 0.7966Epoch 00100: val_loss did not improve
1712/1712 [==============================] - 2s 1ms/step - loss: 0.0010 - acc: 0.7973 - val_loss: 0.0014 - val_acc: 0.7664


 2017-12-11 23:26:13.728801
Using optimizer:  Adadelta
Did not load old weights...
Train on 1712 samples, validate on 428 samples
Epoch 1/100
1696/1712 [============================>.] - ETA: 0s - loss: 9.4268e-04 - acc: 0.8037Epoch 00001: val_loss improved from inf to 0.00136, saving model to weights/modelweights_Adadelta.h5
1712/1712 [==============================] - 2s 1ms/step - loss: 9.4136e-04 - acc: 0.8043 - val_loss: 0.0014 - val_acc: 0.7640
Epoch 2/100
1696/1712 [============================>.] - ETA: 0s - loss: 9.1635e-04 - acc: 0.8078Epoch 00002: val_loss improved from 0.00136 to 0.00135, saving model to weights/modelweights_Adadelta.h5
1712/1712 [==============================] - 2s 1ms/step - loss: 9.1491e-04 - acc: 0.8084 - val_loss: 0.0013 - val_acc: 0.7593
Epoch 3/100
1696/1712 [============================>.] - ETA: 0s - loss: 9.0305e-04 - acc: 0.8007Epoch 00003: val_loss improved from 0.00135 to 0.00135, saving model to weights/modelweights_Adadelta.h5
1712/1712 [==============================] - 2s 1ms/step - loss: 9.0285e-04 - acc: 0.7985 - val_loss: 0.0013 - val_acc: 0.7593
Epoch 4/100
1696/1712 [============================>.] - ETA: 0s - loss: 9.1097e-04 - acc: 0.8025Epoch 00004: val_loss improved from 0.00135 to 0.00134, saving model to weights/modelweights_Adadelta.h5
1712/1712 [==============================] - 2s 1ms/step - loss: 9.1123e-04 - acc: 0.8037 - val_loss: 0.0013 - val_acc: 0.7570
Epoch 5/100
1696/1712 [============================>.] - ETA: 0s - loss: 9.1726e-04 - acc: 0.7871Epoch 00005: val_loss did not improve
1712/1712 [==============================] - 2s 1ms/step - loss: 9.1777e-04 - acc: 0.7862 - val_loss: 0.0013 - val_acc: 0.7593
Epoch 6/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.9849e-04 - acc: 0.8048Epoch 00006: val_loss did not improve
1712/1712 [==============================] - 2s 1ms/step - loss: 9.0448e-04 - acc: 0.8020 - val_loss: 0.0013 - val_acc: 0.7593
Epoch 7/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.8938e-04 - acc: 0.8048Epoch 00007: val_loss did not improve
1712/1712 [==============================] - 2s 1ms/step - loss: 8.9028e-04 - acc: 0.8067 - val_loss: 0.0013 - val_acc: 0.7570
Epoch 8/100
1696/1712 [============================>.] - ETA: 0s - loss: 9.0546e-04 - acc: 0.7972Epoch 00008: val_loss did not improve
1712/1712 [==============================] - 2s 1ms/step - loss: 9.0493e-04 - acc: 0.7967 - val_loss: 0.0013 - val_acc: 0.7570
Epoch 9/100
1696/1712 [============================>.] - ETA: 0s - loss: 9.1433e-04 - acc: 0.8101Epoch 00009: val_loss did not improve
1712/1712 [==============================] - 2s 1ms/step - loss: 9.1336e-04 - acc: 0.8096 - val_loss: 0.0013 - val_acc: 0.7593
Epoch 10/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.9967e-04 - acc: 0.8001Epoch 00010: val_loss improved from 0.00134 to 0.00134, saving model to weights/modelweights_Adadelta.h5
1712/1712 [==============================] - 2s 1ms/step - loss: 8.9918e-04 - acc: 0.7991 - val_loss: 0.0013 - val_acc: 0.7547
Epoch 11/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.9555e-04 - acc: 0.8125Epoch 00011: val_loss did not improve
1712/1712 [==============================] - 2s 1ms/step - loss: 8.9508e-04 - acc: 0.8113 - val_loss: 0.0013 - val_acc: 0.7617
Epoch 12/100
1696/1712 [============================>.] - ETA: 0s - loss: 9.0531e-04 - acc: 0.7983Epoch 00012: val_loss improved from 0.00134 to 0.00133, saving model to weights/modelweights_Adadelta.h5
1712/1712 [==============================] - 2s 1ms/step - loss: 9.0896e-04 - acc: 0.7985 - val_loss: 0.0013 - val_acc: 0.7523
Epoch 13/100
1696/1712 [============================>.] - ETA: 0s - loss: 9.1209e-04 - acc: 0.8031Epoch 00013: val_loss improved from 0.00133 to 0.00132, saving model to weights/modelweights_Adadelta.h5
1712/1712 [==============================] - 2s 1ms/step - loss: 9.1253e-04 - acc: 0.8032 - val_loss: 0.0013 - val_acc: 0.7453
Epoch 14/100
1696/1712 [============================>.] - ETA: 0s - loss: 9.0803e-04 - acc: 0.8037Epoch 00014: val_loss improved from 0.00132 to 0.00132, saving model to weights/modelweights_Adadelta.h5
1712/1712 [==============================] - 2s 1ms/step - loss: 9.0695e-04 - acc: 0.8049 - val_loss: 0.0013 - val_acc: 0.7523
Epoch 15/100
1696/1712 [============================>.] - ETA: 0s - loss: 9.1262e-04 - acc: 0.8048Epoch 00015: val_loss did not improve
1712/1712 [==============================] - 2s 1ms/step - loss: 9.0959e-04 - acc: 0.8049 - val_loss: 0.0013 - val_acc: 0.7617
Epoch 16/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.9673e-04 - acc: 0.7989Epoch 00016: val_loss did not improve
1712/1712 [==============================] - 2s 1ms/step - loss: 8.9725e-04 - acc: 0.7996 - val_loss: 0.0013 - val_acc: 0.7593
Epoch 17/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.9280e-04 - acc: 0.7983Epoch 00017: val_loss improved from 0.00132 to 0.00132, saving model to weights/modelweights_Adadelta.h5
1712/1712 [==============================] - 2s 1ms/step - loss: 8.9591e-04 - acc: 0.7985 - val_loss: 0.0013 - val_acc: 0.7593
Epoch 18/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.7963e-04 - acc: 0.8125Epoch 00018: val_loss did not improve
1712/1712 [==============================] - 2s 1ms/step - loss: 8.7925e-04 - acc: 0.8131 - val_loss: 0.0013 - val_acc: 0.7523
Epoch 19/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.9193e-04 - acc: 0.8013Epoch 00019: val_loss improved from 0.00132 to 0.00131, saving model to weights/modelweights_Adadelta.h5
1712/1712 [==============================] - 2s 1ms/step - loss: 8.9335e-04 - acc: 0.8008 - val_loss: 0.0013 - val_acc: 0.7547
Epoch 20/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.8489e-04 - acc: 0.7942Epoch 00020: val_loss did not improve
1712/1712 [==============================] - 2s 1ms/step - loss: 8.8506e-04 - acc: 0.7956 - val_loss: 0.0013 - val_acc: 0.7477
Epoch 21/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.7191e-04 - acc: 0.8019Epoch 00021: val_loss did not improve
1712/1712 [==============================] - 2s 1ms/step - loss: 8.7335e-04 - acc: 0.8020 - val_loss: 0.0013 - val_acc: 0.7523
Epoch 22/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.7916e-04 - acc: 0.7995Epoch 00022: val_loss did not improve
1712/1712 [==============================] - 2s 1ms/step - loss: 8.7911e-04 - acc: 0.8002 - val_loss: 0.0013 - val_acc: 0.7617
Epoch 23/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.9517e-04 - acc: 0.8025Epoch 00023: val_loss did not improve
1712/1712 [==============================] - 2s 1ms/step - loss: 8.9335e-04 - acc: 0.8037 - val_loss: 0.0013 - val_acc: 0.7547
Epoch 24/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.7764e-04 - acc: 0.7989Epoch 00024: val_loss did not improve
1712/1712 [==============================] - 2s 1ms/step - loss: 8.7948e-04 - acc: 0.7985 - val_loss: 0.0013 - val_acc: 0.7570
Epoch 25/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.6786e-04 - acc: 0.8149Epoch 00025: val_loss did not improve
1712/1712 [==============================] - 2s 1ms/step - loss: 8.7020e-04 - acc: 0.8137 - val_loss: 0.0013 - val_acc: 0.7547
Epoch 26/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.7771e-04 - acc: 0.7960Epoch 00026: val_loss did not improve
1712/1712 [==============================] - 2s 1ms/step - loss: 8.8186e-04 - acc: 0.7961 - val_loss: 0.0013 - val_acc: 0.7570
Epoch 27/100
1696/1712 [============================>.] - ETA: 0s - loss: 9.0278e-04 - acc: 0.7925Epoch 00027: val_loss did not improve
1712/1712 [==============================] - 2s 1ms/step - loss: 9.0317e-04 - acc: 0.7915 - val_loss: 0.0013 - val_acc: 0.7547
Epoch 28/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.7892e-04 - acc: 0.8025Epoch 00028: val_loss did not improve
1712/1712 [==============================] - 2s 1ms/step - loss: 8.7867e-04 - acc: 0.8026 - val_loss: 0.0013 - val_acc: 0.7617
Epoch 29/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.9215e-04 - acc: 0.7989Epoch 00029: val_loss did not improve
1712/1712 [==============================] - 2s 1ms/step - loss: 8.9207e-04 - acc: 0.7991 - val_loss: 0.0013 - val_acc: 0.7523
Epoch 30/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.8736e-04 - acc: 0.8066Epoch 00030: val_loss improved from 0.00131 to 0.00131, saving model to weights/modelweights_Adadelta.h5
1712/1712 [==============================] - 2s 1ms/step - loss: 8.8936e-04 - acc: 0.8067 - val_loss: 0.0013 - val_acc: 0.7570
Epoch 31/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.7811e-04 - acc: 0.7954Epoch 00031: val_loss improved from 0.00131 to 0.00131, saving model to weights/modelweights_Adadelta.h5
1712/1712 [==============================] - 2s 1ms/step - loss: 8.7886e-04 - acc: 0.7961 - val_loss: 0.0013 - val_acc: 0.7570
Epoch 32/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.7753e-04 - acc: 0.8072Epoch 00032: val_loss did not improve
1712/1712 [==============================] - 2s 1ms/step - loss: 8.7557e-04 - acc: 0.8090 - val_loss: 0.0013 - val_acc: 0.7523
Epoch 33/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.9240e-04 - acc: 0.7983Epoch 00033: val_loss did not improve
1712/1712 [==============================] - 2s 1ms/step - loss: 8.9210e-04 - acc: 0.7991 - val_loss: 0.0013 - val_acc: 0.7593
Epoch 34/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.7595e-04 - acc: 0.7966Epoch 00034: val_loss did not improve
1712/1712 [==============================] - 2s 1ms/step - loss: 8.7801e-04 - acc: 0.7973 - val_loss: 0.0013 - val_acc: 0.7570
Epoch 35/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.7467e-04 - acc: 0.8078Epoch 00035: val_loss did not improve
1712/1712 [==============================] - 2s 1ms/step - loss: 8.7521e-04 - acc: 0.8067 - val_loss: 0.0013 - val_acc: 0.7617
Epoch 36/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.9313e-04 - acc: 0.8037Epoch 00036: val_loss did not improve
1712/1712 [==============================] - 2s 1ms/step - loss: 8.9439e-04 - acc: 0.8026 - val_loss: 0.0013 - val_acc: 0.7617
Epoch 37/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.7472e-04 - acc: 0.8113Epoch 00037: val_loss did not improve
1712/1712 [==============================] - 2s 1ms/step - loss: 8.7582e-04 - acc: 0.8113 - val_loss: 0.0013 - val_acc: 0.7523
Epoch 38/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.8127e-04 - acc: 0.8048Epoch 00038: val_loss did not improve
1712/1712 [==============================] - 2s 1ms/step - loss: 8.8069e-04 - acc: 0.8032 - val_loss: 0.0013 - val_acc: 0.7640
Epoch 39/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.7253e-04 - acc: 0.7995Epoch 00039: val_loss did not improve
1712/1712 [==============================] - 2s 1ms/step - loss: 8.7784e-04 - acc: 0.7996 - val_loss: 0.0013 - val_acc: 0.7593
Epoch 40/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.7750e-04 - acc: 0.8190Epoch 00040: val_loss did not improve
1712/1712 [==============================] - 2s 1ms/step - loss: 8.7757e-04 - acc: 0.8178 - val_loss: 0.0013 - val_acc: 0.7593
Epoch 41/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.8046e-04 - acc: 0.8084Epoch 00041: val_loss did not improve
1712/1712 [==============================] - 2s 1ms/step - loss: 8.8008e-04 - acc: 0.8096 - val_loss: 0.0013 - val_acc: 0.7593
Epoch 42/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.7665e-04 - acc: 0.8107Epoch 00042: val_loss did not improve
1712/1712 [==============================] - 2s 1ms/step - loss: 8.7430e-04 - acc: 0.8119 - val_loss: 0.0013 - val_acc: 0.7593
Epoch 43/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.7984e-04 - acc: 0.7907Epoch 00043: val_loss did not improve
1712/1712 [==============================] - 2s 1ms/step - loss: 8.8343e-04 - acc: 0.7915 - val_loss: 0.0013 - val_acc: 0.7617
Epoch 44/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.8186e-04 - acc: 0.8031Epoch 00044: val_loss did not improve
1712/1712 [==============================] - 2s 1ms/step - loss: 8.8381e-04 - acc: 0.8020 - val_loss: 0.0013 - val_acc: 0.7547
Epoch 45/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.6726e-04 - acc: 0.8072Epoch 00045: val_loss did not improve
1712/1712 [==============================] - 2s 1ms/step - loss: 8.7075e-04 - acc: 0.8072 - val_loss: 0.0013 - val_acc: 0.7640
Epoch 46/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.7002e-04 - acc: 0.8001Epoch 00046: val_loss did not improve
1712/1712 [==============================] - 2s 1ms/step - loss: 8.7167e-04 - acc: 0.7991 - val_loss: 0.0013 - val_acc: 0.7570
Epoch 47/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.8403e-04 - acc: 0.8025Epoch 00047: val_loss did not improve
1712/1712 [==============================] - 2s 1ms/step - loss: 8.8302e-04 - acc: 0.8032 - val_loss: 0.0013 - val_acc: 0.7547
Epoch 48/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.7485e-04 - acc: 0.8066Epoch 00048: val_loss did not improve
1712/1712 [==============================] - 2s 1ms/step - loss: 8.7295e-04 - acc: 0.8072 - val_loss: 0.0013 - val_acc: 0.7570
Epoch 49/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.8386e-04 - acc: 0.8007Epoch 00049: val_loss did not improve
1712/1712 [==============================] - 2s 1ms/step - loss: 8.8263e-04 - acc: 0.8020 - val_loss: 0.0013 - val_acc: 0.7547
Epoch 50/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.6968e-04 - acc: 0.8060Epoch 00050: val_loss did not improve
1712/1712 [==============================] - 2s 1ms/step - loss: 8.7036e-04 - acc: 0.8055 - val_loss: 0.0013 - val_acc: 0.7500
Epoch 51/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.6892e-04 - acc: 0.7978Epoch 00051: val_loss did not improve
1712/1712 [==============================] - 2s 1ms/step - loss: 8.6717e-04 - acc: 0.7973 - val_loss: 0.0013 - val_acc: 0.7523
Epoch 52/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.8895e-04 - acc: 0.8031Epoch 00052: val_loss did not improve
1712/1712 [==============================] - 2s 1ms/step - loss: 8.9616e-04 - acc: 0.8032 - val_loss: 0.0013 - val_acc: 0.7570
Epoch 53/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.7750e-04 - acc: 0.8054Epoch 00053: val_loss did not improve
1712/1712 [==============================] - 2s 1ms/step - loss: 8.7452e-04 - acc: 0.8061 - val_loss: 0.0013 - val_acc: 0.7500
Epoch 54/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.8072e-04 - acc: 0.8054Epoch 00054: val_loss did not improve
1712/1712 [==============================] - 2s 1ms/step - loss: 8.7792e-04 - acc: 0.8067 - val_loss: 0.0013 - val_acc: 0.7570
Epoch 55/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.8219e-04 - acc: 0.7942Epoch 00055: val_loss did not improve
1712/1712 [==============================] - 2s 1ms/step - loss: 8.8267e-04 - acc: 0.7938 - val_loss: 0.0013 - val_acc: 0.7500
Epoch 56/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.7800e-04 - acc: 0.8078Epoch 00056: val_loss did not improve
1712/1712 [==============================] - 2s 1ms/step - loss: 8.7894e-04 - acc: 0.8084 - val_loss: 0.0013 - val_acc: 0.7570
Epoch 57/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.6896e-04 - acc: 0.8143Epoch 00057: val_loss did not improve
1712/1712 [==============================] - 2s 1ms/step - loss: 8.7066e-04 - acc: 0.8148 - val_loss: 0.0013 - val_acc: 0.7547
Epoch 58/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.8115e-04 - acc: 0.8013Epoch 00058: val_loss did not improve
1712/1712 [==============================] - 2s 1ms/step - loss: 8.8016e-04 - acc: 0.8008 - val_loss: 0.0013 - val_acc: 0.7500
Epoch 59/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.8054e-04 - acc: 0.7966Epoch 00059: val_loss did not improve
1712/1712 [==============================] - 2s 1ms/step - loss: 8.7942e-04 - acc: 0.7967 - val_loss: 0.0013 - val_acc: 0.7570
Epoch 60/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.8385e-04 - acc: 0.7995Epoch 00060: val_loss did not improve
1712/1712 [==============================] - 2s 1ms/step - loss: 8.8345e-04 - acc: 0.7996 - val_loss: 0.0013 - val_acc: 0.7570
Epoch 61/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.7563e-04 - acc: 0.8013Epoch 00061: val_loss did not improve
1712/1712 [==============================] - 2s 1ms/step - loss: 8.7549e-04 - acc: 0.8020 - val_loss: 0.0013 - val_acc: 0.7500
Epoch 62/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.8784e-04 - acc: 0.7936Epoch 00062: val_loss did not improve
1712/1712 [==============================] - 2s 1ms/step - loss: 8.8501e-04 - acc: 0.7926 - val_loss: 0.0013 - val_acc: 0.7430
Epoch 63/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.8724e-04 - acc: 0.7854Epoch 00063: val_loss did not improve
1712/1712 [==============================] - 2s 1ms/step - loss: 8.8491e-04 - acc: 0.7856 - val_loss: 0.0013 - val_acc: 0.7523
Epoch 64/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.7044e-04 - acc: 0.8090Epoch 00064: val_loss did not improve
1712/1712 [==============================] - 2s 1ms/step - loss: 8.7087e-04 - acc: 0.8078 - val_loss: 0.0013 - val_acc: 0.7523
Epoch 65/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.7283e-04 - acc: 0.7983Epoch 00065: val_loss improved from 0.00131 to 0.00130, saving model to weights/modelweights_Adadelta.h5
1712/1712 [==============================] - 2s 1ms/step - loss: 8.7244e-04 - acc: 0.7991 - val_loss: 0.0013 - val_acc: 0.7523
Epoch 66/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.9246e-04 - acc: 0.8025Epoch 00066: val_loss did not improve
1712/1712 [==============================] - 2s 1ms/step - loss: 8.9023e-04 - acc: 0.8043 - val_loss: 0.0013 - val_acc: 0.7407
Epoch 67/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.8827e-04 - acc: 0.8001Epoch 00067: val_loss did not improve
1712/1712 [==============================] - 2s 1ms/step - loss: 8.9009e-04 - acc: 0.7996 - val_loss: 0.0013 - val_acc: 0.7570
Epoch 68/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.6769e-04 - acc: 0.8072Epoch 00068: val_loss did not improve
1712/1712 [==============================] - 2s 1ms/step - loss: 8.6501e-04 - acc: 0.8078 - val_loss: 0.0013 - val_acc: 0.7547
Epoch 69/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.7150e-04 - acc: 0.7972Epoch 00069: val_loss did not improve
1712/1712 [==============================] - 2s 1ms/step - loss: 8.7196e-04 - acc: 0.7938 - val_loss: 0.0013 - val_acc: 0.7570
Epoch 70/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.7142e-04 - acc: 0.8025Epoch 00070: val_loss did not improve
1712/1712 [==============================] - 2s 1ms/step - loss: 8.7001e-04 - acc: 0.8026 - val_loss: 0.0013 - val_acc: 0.7593
Epoch 71/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.6748e-04 - acc: 0.8090Epoch 00071: val_loss did not improve
1712/1712 [==============================] - 2s 1ms/step - loss: 8.7061e-04 - acc: 0.8084 - val_loss: 0.0013 - val_acc: 0.7570
Epoch 72/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.6391e-04 - acc: 0.8090Epoch 00072: val_loss did not improve
1712/1712 [==============================] - 2s 1ms/step - loss: 8.6265e-04 - acc: 0.8096 - val_loss: 0.0013 - val_acc: 0.7547
Epoch 73/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.8173e-04 - acc: 0.8037Epoch 00073: val_loss did not improve
1712/1712 [==============================] - 2s 1ms/step - loss: 8.8275e-04 - acc: 0.8049 - val_loss: 0.0013 - val_acc: 0.7500
Epoch 74/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.7133e-04 - acc: 0.8101Epoch 00074: val_loss did not improve
1712/1712 [==============================] - 2s 1ms/step - loss: 8.6942e-04 - acc: 0.8119 - val_loss: 0.0013 - val_acc: 0.7570
Epoch 75/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.8266e-04 - acc: 0.7978Epoch 00075: val_loss did not improve
1712/1712 [==============================] - 2s 1ms/step - loss: 8.8430e-04 - acc: 0.7985 - val_loss: 0.0013 - val_acc: 0.7523
Epoch 76/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.7076e-04 - acc: 0.8031Epoch 00076: val_loss did not improve
1712/1712 [==============================] - 2s 1ms/step - loss: 8.7109e-04 - acc: 0.8032 - val_loss: 0.0013 - val_acc: 0.7593
Epoch 77/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.7537e-04 - acc: 0.8048Epoch 00077: val_loss did not improve
1712/1712 [==============================] - 2s 1ms/step - loss: 8.7461e-04 - acc: 0.8049 - val_loss: 0.0013 - val_acc: 0.7617
Epoch 78/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.7254e-04 - acc: 0.8113Epoch 00078: val_loss did not improve
1712/1712 [==============================] - 2s 1ms/step - loss: 8.7459e-04 - acc: 0.8131 - val_loss: 0.0013 - val_acc: 0.7547
Epoch 79/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.7622e-04 - acc: 0.8001Epoch 00079: val_loss did not improve
1712/1712 [==============================] - 2s 1ms/step - loss: 8.7673e-04 - acc: 0.7979 - val_loss: 0.0013 - val_acc: 0.7570
Epoch 80/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.5029e-04 - acc: 0.8131Epoch 00080: val_loss did not improve
1712/1712 [==============================] - 2s 1ms/step - loss: 8.5037e-04 - acc: 0.8113 - val_loss: 0.0013 - val_acc: 0.7570
Epoch 81/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.6623e-04 - acc: 0.8096Epoch 00081: val_loss did not improve
1712/1712 [==============================] - 2s 1ms/step - loss: 8.6661e-04 - acc: 0.8084 - val_loss: 0.0013 - val_acc: 0.7547
Epoch 82/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.8089e-04 - acc: 0.8101Epoch 00082: val_loss did not improve
1712/1712 [==============================] - 2s 1ms/step - loss: 8.8027e-04 - acc: 0.8107 - val_loss: 0.0013 - val_acc: 0.7547
Epoch 83/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.8370e-04 - acc: 0.8096Epoch 00083: val_loss did not improve
1712/1712 [==============================] - 2s 1ms/step - loss: 8.8228e-04 - acc: 0.8113 - val_loss: 0.0013 - val_acc: 0.7547
Epoch 84/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.5470e-04 - acc: 0.8096Epoch 00084: val_loss did not improve
1712/1712 [==============================] - 2s 1ms/step - loss: 8.5432e-04 - acc: 0.8102 - val_loss: 0.0013 - val_acc: 0.7593
Epoch 85/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.6364e-04 - acc: 0.8131Epoch 00085: val_loss did not improve
1712/1712 [==============================] - 2s 1ms/step - loss: 8.6763e-04 - acc: 0.8137 - val_loss: 0.0013 - val_acc: 0.7570
Epoch 86/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.7391e-04 - acc: 0.8137Epoch 00086: val_loss did not improve
1712/1712 [==============================] - 2s 1ms/step - loss: 8.7544e-04 - acc: 0.8143 - val_loss: 0.0013 - val_acc: 0.7570
Epoch 87/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.6100e-04 - acc: 0.8190Epoch 00087: val_loss did not improve
1712/1712 [==============================] - 2s 1ms/step - loss: 8.6341e-04 - acc: 0.8183 - val_loss: 0.0013 - val_acc: 0.7523
Epoch 88/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.6029e-04 - acc: 0.7978Epoch 00088: val_loss did not improve
1712/1712 [==============================] - 2s 1ms/step - loss: 8.6038e-04 - acc: 0.7967 - val_loss: 0.0013 - val_acc: 0.7430
Epoch 89/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.6849e-04 - acc: 0.8090Epoch 00089: val_loss did not improve
1712/1712 [==============================] - 2s 1ms/step - loss: 8.7100e-04 - acc: 0.8090 - val_loss: 0.0013 - val_acc: 0.7547
Epoch 90/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.5492e-04 - acc: 0.7989Epoch 00090: val_loss did not improve
1712/1712 [==============================] - 2s 1ms/step - loss: 8.5435e-04 - acc: 0.7973 - val_loss: 0.0013 - val_acc: 0.7593
Epoch 91/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.7021e-04 - acc: 0.8060Epoch 00091: val_loss did not improve
1712/1712 [==============================] - 2s 1ms/step - loss: 8.7609e-04 - acc: 0.8055 - val_loss: 0.0013 - val_acc: 0.7593
Epoch 92/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.7720e-04 - acc: 0.7960Epoch 00092: val_loss did not improve
1712/1712 [==============================] - 2s 1ms/step - loss: 8.7794e-04 - acc: 0.7956 - val_loss: 0.0013 - val_acc: 0.7570
Epoch 93/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.7586e-04 - acc: 0.8072Epoch 00093: val_loss did not improve
1712/1712 [==============================] - 2s 1ms/step - loss: 8.7721e-04 - acc: 0.8061 - val_loss: 0.0013 - val_acc: 0.7500
Epoch 94/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.6512e-04 - acc: 0.8025Epoch 00094: val_loss did not improve
1712/1712 [==============================] - 2s 1ms/step - loss: 8.6447e-04 - acc: 0.8026 - val_loss: 0.0013 - val_acc: 0.7547
Epoch 95/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.7011e-04 - acc: 0.8131Epoch 00095: val_loss improved from 0.00130 to 0.00130, saving model to weights/modelweights_Adadelta.h5
1712/1712 [==============================] - 2s 1ms/step - loss: 8.6998e-04 - acc: 0.8125 - val_loss: 0.0013 - val_acc: 0.7500
Epoch 96/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.6861e-04 - acc: 0.8154Epoch 00096: val_loss did not improve
1712/1712 [==============================] - 2s 1ms/step - loss: 8.6855e-04 - acc: 0.8154 - val_loss: 0.0013 - val_acc: 0.7500
Epoch 97/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.6096e-04 - acc: 0.8119Epoch 00097: val_loss did not improve
1712/1712 [==============================] - 2s 1ms/step - loss: 8.6126e-04 - acc: 0.8125 - val_loss: 0.0013 - val_acc: 0.7570
Epoch 98/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.6197e-04 - acc: 0.8019Epoch 00098: val_loss did not improve
1712/1712 [==============================] - 2s 1ms/step - loss: 8.6151e-04 - acc: 0.8020 - val_loss: 0.0013 - val_acc: 0.7523
Epoch 99/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.6081e-04 - acc: 0.7930Epoch 00099: val_loss did not improve
1712/1712 [==============================] - 2s 1ms/step - loss: 8.6103e-04 - acc: 0.7926 - val_loss: 0.0013 - val_acc: 0.7477
Epoch 100/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.7162e-04 - acc: 0.8001Epoch 00100: val_loss did not improve
1712/1712 [==============================] - 2s 1ms/step - loss: 8.6953e-04 - acc: 0.8014 - val_loss: 0.0013 - val_acc: 0.7523


 2017-12-11 23:29:17.297930
Using optimizer:  Adamax
Did not load old weights...
Train on 1712 samples, validate on 428 samples
Epoch 1/100
1696/1712 [============================>.] - ETA: 0s - loss: 9.2794e-04 - acc: 0.7995Epoch 00001: val_loss improved from inf to 0.00130, saving model to weights/modelweights_Adamax.h5
1712/1712 [==============================] - 2s 1ms/step - loss: 9.2970e-04 - acc: 0.8002 - val_loss: 0.0013 - val_acc: 0.7570
Epoch 2/100
1696/1712 [============================>.] - ETA: 0s - loss: 9.0428e-04 - acc: 0.7983Epoch 00002: val_loss did not improve
1712/1712 [==============================] - 2s 989us/step - loss: 9.0499e-04 - acc: 0.7985 - val_loss: 0.0013 - val_acc: 0.7430
Epoch 3/100
1696/1712 [============================>.] - ETA: 0s - loss: 9.1453e-04 - acc: 0.7925Epoch 00003: val_loss did not improve
1712/1712 [==============================] - 2s 986us/step - loss: 9.1372e-04 - acc: 0.7921 - val_loss: 0.0013 - val_acc: 0.7617
Epoch 4/100
1696/1712 [============================>.] - ETA: 0s - loss: 9.1374e-04 - acc: 0.8107Epoch 00004: val_loss did not improve
1712/1712 [==============================] - 2s 984us/step - loss: 9.1390e-04 - acc: 0.8107 - val_loss: 0.0014 - val_acc: 0.7687
Epoch 5/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.9768e-04 - acc: 0.7942Epoch 00005: val_loss did not improve
1712/1712 [==============================] - 2s 982us/step - loss: 8.9814e-04 - acc: 0.7944 - val_loss: 0.0013 - val_acc: 0.7547
Epoch 6/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.9515e-04 - acc: 0.7983Epoch 00006: val_loss did not improve
1712/1712 [==============================] - 2s 985us/step - loss: 8.9515e-04 - acc: 0.7967 - val_loss: 0.0013 - val_acc: 0.7687
Epoch 7/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.9277e-04 - acc: 0.7960Epoch 00007: val_loss did not improve
1712/1712 [==============================] - 2s 987us/step - loss: 8.9187e-04 - acc: 0.7967 - val_loss: 0.0013 - val_acc: 0.7453
Epoch 8/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.8602e-04 - acc: 0.7954Epoch 00008: val_loss did not improve
1712/1712 [==============================] - 2s 984us/step - loss: 8.8644e-04 - acc: 0.7961 - val_loss: 0.0013 - val_acc: 0.7360
Epoch 9/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.8931e-04 - acc: 0.7972Epoch 00009: val_loss did not improve
1712/1712 [==============================] - 2s 986us/step - loss: 8.9090e-04 - acc: 0.7973 - val_loss: 0.0013 - val_acc: 0.7570
Epoch 10/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.8055e-04 - acc: 0.8078Epoch 00010: val_loss did not improve
1712/1712 [==============================] - 2s 983us/step - loss: 8.8102e-04 - acc: 0.8090 - val_loss: 0.0013 - val_acc: 0.7640
Epoch 11/100
1696/1712 [============================>.] - ETA: 0s - loss: 9.0129e-04 - acc: 0.8025Epoch 00011: val_loss did not improve
1712/1712 [==============================] - 2s 985us/step - loss: 9.0184e-04 - acc: 0.8014 - val_loss: 0.0013 - val_acc: 0.7593
Epoch 12/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.6697e-04 - acc: 0.8060Epoch 00012: val_loss did not improve
1712/1712 [==============================] - 2s 991us/step - loss: 8.6523e-04 - acc: 0.8043 - val_loss: 0.0013 - val_acc: 0.7500
Epoch 13/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.8269e-04 - acc: 0.8084Epoch 00013: val_loss did not improve
1712/1712 [==============================] - 2s 989us/step - loss: 8.8149e-04 - acc: 0.8067 - val_loss: 0.0013 - val_acc: 0.7453
Epoch 14/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.8221e-04 - acc: 0.8143Epoch 00014: val_loss did not improve
1712/1712 [==============================] - 2s 986us/step - loss: 8.8422e-04 - acc: 0.8137 - val_loss: 0.0013 - val_acc: 0.7710
Epoch 15/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.9085e-04 - acc: 0.8143Epoch 00015: val_loss did not improve
1712/1712 [==============================] - 2s 985us/step - loss: 8.9141e-04 - acc: 0.8148 - val_loss: 0.0013 - val_acc: 0.7477
Epoch 16/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.9582e-04 - acc: 0.8113Epoch 00016: val_loss improved from 0.00130 to 0.00130, saving model to weights/modelweights_Adamax.h5
1712/1712 [==============================] - 2s 1ms/step - loss: 8.9538e-04 - acc: 0.8119 - val_loss: 0.0013 - val_acc: 0.7664
Epoch 17/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.7241e-04 - acc: 0.8042Epoch 00017: val_loss did not improve
1712/1712 [==============================] - 2s 991us/step - loss: 8.7461e-04 - acc: 0.8043 - val_loss: 0.0013 - val_acc: 0.7453
Epoch 18/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.6236e-04 - acc: 0.8137Epoch 00018: val_loss did not improve
1712/1712 [==============================] - 2s 987us/step - loss: 8.6837e-04 - acc: 0.8137 - val_loss: 0.0013 - val_acc: 0.7640
Epoch 19/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.7913e-04 - acc: 0.7978Epoch 00019: val_loss improved from 0.00130 to 0.00130, saving model to weights/modelweights_Adamax.h5
1712/1712 [==============================] - 2s 1ms/step - loss: 8.8010e-04 - acc: 0.7985 - val_loss: 0.0013 - val_acc: 0.7547
Epoch 20/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.5812e-04 - acc: 0.8048Epoch 00020: val_loss did not improve
1712/1712 [==============================] - 2s 985us/step - loss: 8.5725e-04 - acc: 0.8061 - val_loss: 0.0013 - val_acc: 0.7430
Epoch 21/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.9180e-04 - acc: 0.8042Epoch 00021: val_loss did not improve
1712/1712 [==============================] - 2s 987us/step - loss: 8.9117e-04 - acc: 0.8049 - val_loss: 0.0013 - val_acc: 0.7593
Epoch 22/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.7286e-04 - acc: 0.7972Epoch 00022: val_loss did not improve
1712/1712 [==============================] - 2s 984us/step - loss: 8.7291e-04 - acc: 0.7985 - val_loss: 0.0013 - val_acc: 0.7523
Epoch 23/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.6733e-04 - acc: 0.8031Epoch 00023: val_loss did not improve
1712/1712 [==============================] - 2s 987us/step - loss: 8.6899e-04 - acc: 0.8026 - val_loss: 0.0013 - val_acc: 0.7640
Epoch 24/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.8590e-04 - acc: 0.8196Epoch 00024: val_loss did not improve
1712/1712 [==============================] - 2s 988us/step - loss: 8.9054e-04 - acc: 0.8189 - val_loss: 0.0013 - val_acc: 0.7523
Epoch 25/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.7392e-04 - acc: 0.7936Epoch 00025: val_loss did not improve
1712/1712 [==============================] - 2s 991us/step - loss: 8.7445e-04 - acc: 0.7932 - val_loss: 0.0013 - val_acc: 0.7547
Epoch 26/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.8247e-04 - acc: 0.8048Epoch 00026: val_loss did not improve
1712/1712 [==============================] - 2s 992us/step - loss: 8.8322e-04 - acc: 0.8037 - val_loss: 0.0013 - val_acc: 0.7687
Epoch 27/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.8037e-04 - acc: 0.7866Epoch 00027: val_loss did not improve
1712/1712 [==============================] - 2s 989us/step - loss: 8.8209e-04 - acc: 0.7862 - val_loss: 0.0013 - val_acc: 0.7570
Epoch 28/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.9283e-04 - acc: 0.8001Epoch 00028: val_loss did not improve
1712/1712 [==============================] - 2s 988us/step - loss: 8.9230e-04 - acc: 0.7973 - val_loss: 0.0013 - val_acc: 0.7477
Epoch 29/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.6245e-04 - acc: 0.8048Epoch 00029: val_loss did not improve
1712/1712 [==============================] - 2s 985us/step - loss: 8.6296e-04 - acc: 0.8037 - val_loss: 0.0013 - val_acc: 0.7664
Epoch 30/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.7043e-04 - acc: 0.8137Epoch 00030: val_loss did not improve
1712/1712 [==============================] - 2s 985us/step - loss: 8.7004e-04 - acc: 0.8119 - val_loss: 0.0013 - val_acc: 0.7664
Epoch 31/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.7699e-04 - acc: 0.8037Epoch 00031: val_loss did not improve
1712/1712 [==============================] - 2s 991us/step - loss: 8.7683e-04 - acc: 0.8043 - val_loss: 0.0013 - val_acc: 0.7570
Epoch 32/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.6090e-04 - acc: 0.8072Epoch 00032: val_loss did not improve
1712/1712 [==============================] - 2s 989us/step - loss: 8.5967e-04 - acc: 0.8061 - val_loss: 0.0013 - val_acc: 0.7617
Epoch 33/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.5311e-04 - acc: 0.8048Epoch 00033: val_loss did not improve
1712/1712 [==============================] - 2s 990us/step - loss: 8.5090e-04 - acc: 0.8049 - val_loss: 0.0013 - val_acc: 0.7664
Epoch 34/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.8293e-04 - acc: 0.8072Epoch 00034: val_loss did not improve
1712/1712 [==============================] - 2s 987us/step - loss: 8.8364e-04 - acc: 0.8072 - val_loss: 0.0013 - val_acc: 0.7547
Epoch 35/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.6635e-04 - acc: 0.8048Epoch 00035: val_loss did not improve
1712/1712 [==============================] - 2s 991us/step - loss: 8.6484e-04 - acc: 0.8055 - val_loss: 0.0013 - val_acc: 0.7664
Epoch 36/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.9174e-04 - acc: 0.8054Epoch 00036: val_loss did not improve
1712/1712 [==============================] - 2s 991us/step - loss: 8.9527e-04 - acc: 0.8049 - val_loss: 0.0013 - val_acc: 0.7617
Epoch 37/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.7162e-04 - acc: 0.8078Epoch 00037: val_loss did not improve
1712/1712 [==============================] - 2s 990us/step - loss: 8.7105e-04 - acc: 0.8078 - val_loss: 0.0013 - val_acc: 0.7500
Epoch 38/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.7676e-04 - acc: 0.7889Epoch 00038: val_loss did not improve
1712/1712 [==============================] - 2s 989us/step - loss: 8.7933e-04 - acc: 0.7891 - val_loss: 0.0013 - val_acc: 0.7593
Epoch 39/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.6697e-04 - acc: 0.8060Epoch 00039: val_loss did not improve
1712/1712 [==============================] - 2s 992us/step - loss: 8.6822e-04 - acc: 0.8061 - val_loss: 0.0013 - val_acc: 0.7547
Epoch 40/100
1696/1712 [============================>.] - ETA: 0s - loss: 9.0446e-04 - acc: 0.8054Epoch 00040: val_loss did not improve
1712/1712 [==============================] - 2s 993us/step - loss: 9.0324e-04 - acc: 0.8067 - val_loss: 0.0013 - val_acc: 0.7570
Epoch 41/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.5351e-04 - acc: 0.8066Epoch 00041: val_loss did not improve
1712/1712 [==============================] - 2s 989us/step - loss: 8.5410e-04 - acc: 0.8072 - val_loss: 0.0013 - val_acc: 0.7547
Epoch 42/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.4447e-04 - acc: 0.8137Epoch 00042: val_loss did not improve
1712/1712 [==============================] - 2s 992us/step - loss: 8.4668e-04 - acc: 0.8148 - val_loss: 0.0013 - val_acc: 0.7734
Epoch 43/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.4648e-04 - acc: 0.8072Epoch 00043: val_loss did not improve
1712/1712 [==============================] - 2s 987us/step - loss: 8.4526e-04 - acc: 0.8055 - val_loss: 0.0013 - val_acc: 0.7617
Epoch 44/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.7134e-04 - acc: 0.8042Epoch 00044: val_loss did not improve
1712/1712 [==============================] - 2s 989us/step - loss: 8.7002e-04 - acc: 0.8043 - val_loss: 0.0013 - val_acc: 0.7523
Epoch 45/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.6310e-04 - acc: 0.8137Epoch 00045: val_loss did not improve
1712/1712 [==============================] - 2s 990us/step - loss: 8.6242e-04 - acc: 0.8125 - val_loss: 0.0013 - val_acc: 0.7617
Epoch 46/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.5812e-04 - acc: 0.7989Epoch 00046: val_loss did not improve
1712/1712 [==============================] - 2s 987us/step - loss: 8.6039e-04 - acc: 0.7979 - val_loss: 0.0014 - val_acc: 0.7593
Epoch 47/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.5306e-04 - acc: 0.8160Epoch 00047: val_loss did not improve
1712/1712 [==============================] - 2s 983us/step - loss: 8.5151e-04 - acc: 0.8172 - val_loss: 0.0013 - val_acc: 0.7500
Epoch 48/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.6362e-04 - acc: 0.7942Epoch 00048: val_loss did not improve
1712/1712 [==============================] - 2s 987us/step - loss: 8.6386e-04 - acc: 0.7938 - val_loss: 0.0013 - val_acc: 0.7570
Epoch 49/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.6304e-04 - acc: 0.8178Epoch 00049: val_loss did not improve
1712/1712 [==============================] - 2s 985us/step - loss: 8.6332e-04 - acc: 0.8178 - val_loss: 0.0013 - val_acc: 0.7547
Epoch 50/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.5766e-04 - acc: 0.8131Epoch 00050: val_loss did not improve
1712/1712 [==============================] - 2s 988us/step - loss: 8.5558e-04 - acc: 0.8113 - val_loss: 0.0013 - val_acc: 0.7570
Epoch 51/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.6300e-04 - acc: 0.8007Epoch 00051: val_loss did not improve
1712/1712 [==============================] - 2s 990us/step - loss: 8.6253e-04 - acc: 0.7991 - val_loss: 0.0013 - val_acc: 0.7547
Epoch 52/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.6051e-04 - acc: 0.8084Epoch 00052: val_loss did not improve
1712/1712 [==============================] - 2s 988us/step - loss: 8.6230e-04 - acc: 0.8084 - val_loss: 0.0013 - val_acc: 0.7640
Epoch 53/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.5780e-04 - acc: 0.8060Epoch 00053: val_loss did not improve
1712/1712 [==============================] - 2s 984us/step - loss: 8.5622e-04 - acc: 0.8061 - val_loss: 0.0013 - val_acc: 0.7570
Epoch 54/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.5317e-04 - acc: 0.7925Epoch 00054: val_loss did not improve
1712/1712 [==============================] - 2s 987us/step - loss: 8.5255e-04 - acc: 0.7926 - val_loss: 0.0013 - val_acc: 0.7687
Epoch 55/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.7127e-04 - acc: 0.8066Epoch 00055: val_loss did not improve
1712/1712 [==============================] - 2s 986us/step - loss: 8.7026e-04 - acc: 0.8072 - val_loss: 0.0013 - val_acc: 0.7617
Epoch 56/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.6139e-04 - acc: 0.8172Epoch 00056: val_loss did not improve
1712/1712 [==============================] - 2s 985us/step - loss: 8.6505e-04 - acc: 0.8166 - val_loss: 0.0013 - val_acc: 0.7664
Epoch 57/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.7712e-04 - acc: 0.8160Epoch 00057: val_loss did not improve
1712/1712 [==============================] - 2s 983us/step - loss: 8.7955e-04 - acc: 0.8166 - val_loss: 0.0013 - val_acc: 0.7640
Epoch 58/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.6075e-04 - acc: 0.7942Epoch 00058: val_loss did not improve
1712/1712 [==============================] - 2s 991us/step - loss: 8.5954e-04 - acc: 0.7932 - val_loss: 0.0013 - val_acc: 0.7664
Epoch 59/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.6606e-04 - acc: 0.8090Epoch 00059: val_loss did not improve
1712/1712 [==============================] - 2s 987us/step - loss: 8.6540e-04 - acc: 0.8078 - val_loss: 0.0013 - val_acc: 0.7477
Epoch 60/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.4962e-04 - acc: 0.8019Epoch 00060: val_loss did not improve
1712/1712 [==============================] - 2s 990us/step - loss: 8.4937e-04 - acc: 0.8020 - val_loss: 0.0014 - val_acc: 0.7570
Epoch 61/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.6882e-04 - acc: 0.8060Epoch 00061: val_loss did not improve
1712/1712 [==============================] - 2s 986us/step - loss: 8.6899e-04 - acc: 0.8043 - val_loss: 0.0013 - val_acc: 0.7547
Epoch 62/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.7364e-04 - acc: 0.8149Epoch 00062: val_loss did not improve
1712/1712 [==============================] - 2s 986us/step - loss: 8.7158e-04 - acc: 0.8143 - val_loss: 0.0013 - val_acc: 0.7453
Epoch 63/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.6812e-04 - acc: 0.8025Epoch 00063: val_loss did not improve
1712/1712 [==============================] - 2s 982us/step - loss: 8.6601e-04 - acc: 0.8020 - val_loss: 0.0013 - val_acc: 0.7407
Epoch 64/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.7223e-04 - acc: 0.8107Epoch 00064: val_loss did not improve
1712/1712 [==============================] - 2s 988us/step - loss: 8.7098e-04 - acc: 0.8113 - val_loss: 0.0013 - val_acc: 0.7617
Epoch 65/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.5683e-04 - acc: 0.8054Epoch 00065: val_loss did not improve
1712/1712 [==============================] - 2s 988us/step - loss: 8.5730e-04 - acc: 0.8061 - val_loss: 0.0013 - val_acc: 0.7640
Epoch 66/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.7248e-04 - acc: 0.7930Epoch 00066: val_loss did not improve
1712/1712 [==============================] - 2s 991us/step - loss: 8.7178e-04 - acc: 0.7932 - val_loss: 0.0013 - val_acc: 0.7687
Epoch 67/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.4583e-04 - acc: 0.8113Epoch 00067: val_loss did not improve
1712/1712 [==============================] - 2s 994us/step - loss: 8.4471e-04 - acc: 0.8113 - val_loss: 0.0013 - val_acc: 0.7547
Epoch 68/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.7319e-04 - acc: 0.8019Epoch 00068: val_loss did not improve
1712/1712 [==============================] - 2s 994us/step - loss: 8.7265e-04 - acc: 0.8008 - val_loss: 0.0013 - val_acc: 0.7617
Epoch 69/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.3647e-04 - acc: 0.8019Epoch 00069: val_loss did not improve
1712/1712 [==============================] - 2s 998us/step - loss: 8.3396e-04 - acc: 0.8032 - val_loss: 0.0013 - val_acc: 0.7500
Epoch 70/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.4539e-04 - acc: 0.8125Epoch 00070: val_loss did not improve
1712/1712 [==============================] - 2s 990us/step - loss: 8.4329e-04 - acc: 0.8125 - val_loss: 0.0013 - val_acc: 0.7547
Epoch 71/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.5389e-04 - acc: 0.8042Epoch 00071: val_loss did not improve
1712/1712 [==============================] - 2s 985us/step - loss: 8.5193e-04 - acc: 0.8049 - val_loss: 0.0013 - val_acc: 0.7640
Epoch 72/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.6081e-04 - acc: 0.8096Epoch 00072: val_loss did not improve
1712/1712 [==============================] - 2s 981us/step - loss: 8.6046e-04 - acc: 0.8096 - val_loss: 0.0013 - val_acc: 0.7664
Epoch 73/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.4960e-04 - acc: 0.8054Epoch 00073: val_loss did not improve
1712/1712 [==============================] - 2s 989us/step - loss: 8.4833e-04 - acc: 0.8072 - val_loss: 0.0014 - val_acc: 0.7617
Epoch 74/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.2555e-04 - acc: 0.8084Epoch 00074: val_loss did not improve
1712/1712 [==============================] - 2s 986us/step - loss: 8.2498e-04 - acc: 0.8096 - val_loss: 0.0013 - val_acc: 0.7640
Epoch 75/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.4003e-04 - acc: 0.8096Epoch 00075: val_loss did not improve
1712/1712 [==============================] - 2s 991us/step - loss: 8.3819e-04 - acc: 0.8107 - val_loss: 0.0013 - val_acc: 0.7593
Epoch 76/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.5755e-04 - acc: 0.7966Epoch 00076: val_loss did not improve
1712/1712 [==============================] - 2s 991us/step - loss: 8.6202e-04 - acc: 0.7967 - val_loss: 0.0014 - val_acc: 0.7617
Epoch 77/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.5730e-04 - acc: 0.8042Epoch 00077: val_loss did not improve
1712/1712 [==============================] - 2s 994us/step - loss: 8.5715e-04 - acc: 0.8037 - val_loss: 0.0013 - val_acc: 0.7477
Epoch 78/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.6566e-04 - acc: 0.8107Epoch 00078: val_loss did not improve
1712/1712 [==============================] - 2s 987us/step - loss: 8.6532e-04 - acc: 0.8107 - val_loss: 0.0013 - val_acc: 0.7547
Epoch 79/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.4636e-04 - acc: 0.7883Epoch 00079: val_loss did not improve
1712/1712 [==============================] - 2s 985us/step - loss: 8.4504e-04 - acc: 0.7886 - val_loss: 0.0013 - val_acc: 0.7547
Epoch 80/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.4440e-04 - acc: 0.8007Epoch 00080: val_loss did not improve
1712/1712 [==============================] - 2s 985us/step - loss: 8.4653e-04 - acc: 0.8002 - val_loss: 0.0013 - val_acc: 0.7523
Epoch 81/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.4760e-04 - acc: 0.8072Epoch 00081: val_loss did not improve
1712/1712 [==============================] - 2s 988us/step - loss: 8.4812e-04 - acc: 0.8072 - val_loss: 0.0013 - val_acc: 0.7617
Epoch 82/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.3544e-04 - acc: 0.8101Epoch 00082: val_loss did not improve
1712/1712 [==============================] - 2s 987us/step - loss: 8.3460e-04 - acc: 0.8102 - val_loss: 0.0013 - val_acc: 0.7593
Epoch 83/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.2177e-04 - acc: 0.8048Epoch 00083: val_loss did not improve
1712/1712 [==============================] - 2s 983us/step - loss: 8.2105e-04 - acc: 0.8055 - val_loss: 0.0013 - val_acc: 0.7593
Epoch 84/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.6117e-04 - acc: 0.8208Epoch 00084: val_loss did not improve
1712/1712 [==============================] - 2s 984us/step - loss: 8.5982e-04 - acc: 0.8207 - val_loss: 0.0013 - val_acc: 0.7430
Epoch 85/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.4692e-04 - acc: 0.8019Epoch 00085: val_loss did not improve
1712/1712 [==============================] - 2s 989us/step - loss: 8.4507e-04 - acc: 0.8014 - val_loss: 0.0013 - val_acc: 0.7617
Epoch 86/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.3747e-04 - acc: 0.8237Epoch 00086: val_loss did not improve
1712/1712 [==============================] - 2s 987us/step - loss: 8.3896e-04 - acc: 0.8213 - val_loss: 0.0013 - val_acc: 0.7617
Epoch 87/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.6217e-04 - acc: 0.8149Epoch 00087: val_loss did not improve
1712/1712 [==============================] - 2s 986us/step - loss: 8.6161e-04 - acc: 0.8148 - val_loss: 0.0013 - val_acc: 0.7734
Epoch 88/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.4874e-04 - acc: 0.7960Epoch 00088: val_loss did not improve
1712/1712 [==============================] - 2s 985us/step - loss: 8.5003e-04 - acc: 0.7956 - val_loss: 0.0013 - val_acc: 0.7640
Epoch 89/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.4201e-04 - acc: 0.8042Epoch 00089: val_loss did not improve
1712/1712 [==============================] - 2s 984us/step - loss: 8.4089e-04 - acc: 0.8037 - val_loss: 0.0013 - val_acc: 0.7477
Epoch 90/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.3664e-04 - acc: 0.8084Epoch 00090: val_loss did not improve
1712/1712 [==============================] - 2s 986us/step - loss: 8.3673e-04 - acc: 0.8096 - val_loss: 0.0013 - val_acc: 0.7687
Epoch 91/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.4874e-04 - acc: 0.8154Epoch 00091: val_loss did not improve
1712/1712 [==============================] - 2s 982us/step - loss: 8.4682e-04 - acc: 0.8143 - val_loss: 0.0013 - val_acc: 0.7617
Epoch 92/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.3133e-04 - acc: 0.8037Epoch 00092: val_loss did not improve
1712/1712 [==============================] - 2s 980us/step - loss: 8.3048e-04 - acc: 0.8037 - val_loss: 0.0013 - val_acc: 0.7640
Epoch 93/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.2743e-04 - acc: 0.8031Epoch 00093: val_loss did not improve
1712/1712 [==============================] - 2s 985us/step - loss: 8.2705e-04 - acc: 0.8037 - val_loss: 0.0013 - val_acc: 0.7664
Epoch 94/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.2621e-04 - acc: 0.8031Epoch 00094: val_loss did not improve
1712/1712 [==============================] - 2s 986us/step - loss: 8.2525e-04 - acc: 0.8049 - val_loss: 0.0013 - val_acc: 0.7780
Epoch 95/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.4197e-04 - acc: 0.8078Epoch 00095: val_loss did not improve
1712/1712 [==============================] - 2s 985us/step - loss: 8.4029e-04 - acc: 0.8072 - val_loss: 0.0013 - val_acc: 0.7687
Epoch 96/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.6439e-04 - acc: 0.8090Epoch 00096: val_loss did not improve
1712/1712 [==============================] - 2s 988us/step - loss: 8.6341e-04 - acc: 0.8096 - val_loss: 0.0013 - val_acc: 0.7687
Epoch 97/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.3623e-04 - acc: 0.8149Epoch 00097: val_loss did not improve
1712/1712 [==============================] - 2s 990us/step - loss: 8.3485e-04 - acc: 0.8166 - val_loss: 0.0013 - val_acc: 0.7640
Epoch 98/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.3836e-04 - acc: 0.7995Epoch 00098: val_loss did not improve
1712/1712 [==============================] - 2s 986us/step - loss: 8.3837e-04 - acc: 0.8002 - val_loss: 0.0013 - val_acc: 0.7617
Epoch 99/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.3287e-04 - acc: 0.8243Epoch 00099: val_loss did not improve
1712/1712 [==============================] - 2s 984us/step - loss: 8.3226e-04 - acc: 0.8230 - val_loss: 0.0013 - val_acc: 0.7570
Epoch 100/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.2888e-04 - acc: 0.8166Epoch 00100: val_loss did not improve
1712/1712 [==============================] - 2s 983us/step - loss: 8.2789e-04 - acc: 0.8172 - val_loss: 0.0013 - val_acc: 0.7640


 2017-12-11 23:32:07.827279
Using optimizer:  Adagrad
Did not load old weights...
Train on 1712 samples, validate on 428 samples
Epoch 1/100
1696/1712 [============================>.] - ETA: 0s - loss: 0.0014 - acc: 0.7712Epoch 00001: val_loss improved from inf to 0.00141, saving model to weights/modelweights_Adagrad.h5
1712/1712 [==============================] - 2s 1ms/step - loss: 0.0014 - acc: 0.7716 - val_loss: 0.0014 - val_acc: 0.7453
Epoch 2/100
1696/1712 [============================>.] - ETA: 0s - loss: 9.5076e-04 - acc: 0.7925Epoch 00002: val_loss improved from 0.00141 to 0.00137, saving model to weights/modelweights_Adagrad.h5
1712/1712 [==============================] - 2s 1ms/step - loss: 9.4826e-04 - acc: 0.7909 - val_loss: 0.0014 - val_acc: 0.7547
Epoch 3/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.8607e-04 - acc: 0.7983Epoch 00003: val_loss improved from 0.00137 to 0.00136, saving model to weights/modelweights_Adagrad.h5
1712/1712 [==============================] - 2s 1ms/step - loss: 8.8526e-04 - acc: 0.8002 - val_loss: 0.0014 - val_acc: 0.7523
Epoch 4/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.8372e-04 - acc: 0.8107- ETA: 1s - loss: 8Epoch 00004: val_loss improved from 0.00136 to 0.00133, saving model to weights/modelweights_Adagrad.h5
1712/1712 [==============================] - 2s 1ms/step - loss: 8.8305e-04 - acc: 0.8113 - val_loss: 0.0013 - val_acc: 0.7640
Epoch 5/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.7865e-04 - acc: 0.8066Epoch 00005: val_loss did not improve
1712/1712 [==============================] - 2s 966us/step - loss: 8.7933e-04 - acc: 0.8072 - val_loss: 0.0013 - val_acc: 0.7500
Epoch 6/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.6352e-04 - acc: 0.8031Epoch 00006: val_loss did not improve
1712/1712 [==============================] - 2s 967us/step - loss: 8.6330e-04 - acc: 0.8026 - val_loss: 0.0014 - val_acc: 0.7640
Epoch 7/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.7068e-04 - acc: 0.8007Epoch 00007: val_loss did not improve
1712/1712 [==============================] - 2s 965us/step - loss: 8.6985e-04 - acc: 0.7991 - val_loss: 0.0013 - val_acc: 0.7570
Epoch 8/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.5911e-04 - acc: 0.8231Epoch 00008: val_loss improved from 0.00133 to 0.00133, saving model to weights/modelweights_Adagrad.h5
1712/1712 [==============================] - 2s 1ms/step - loss: 8.6156e-04 - acc: 0.8236 - val_loss: 0.0013 - val_acc: 0.7617
Epoch 9/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.6294e-04 - acc: 0.8113Epoch 00009: val_loss did not improve
1712/1712 [==============================] - 2s 970us/step - loss: 8.6260e-04 - acc: 0.8119 - val_loss: 0.0014 - val_acc: 0.7617
Epoch 10/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.4606e-04 - acc: 0.8160Epoch 00010: val_loss improved from 0.00133 to 0.00133, saving model to weights/modelweights_Adagrad.h5
1712/1712 [==============================] - 2s 1ms/step - loss: 8.4385e-04 - acc: 0.8160 - val_loss: 0.0013 - val_acc: 0.7547
Epoch 11/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.3585e-04 - acc: 0.8101Epoch 00011: val_loss did not improve
1712/1712 [==============================] - 2s 969us/step - loss: 8.3912e-04 - acc: 0.8102 - val_loss: 0.0013 - val_acc: 0.7570
Epoch 12/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.1807e-04 - acc: 0.8037Epoch 00012: val_loss did not improve
1712/1712 [==============================] - 2s 968us/step - loss: 8.1723e-04 - acc: 0.8037 - val_loss: 0.0013 - val_acc: 0.7593
Epoch 13/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.2371e-04 - acc: 0.8184Epoch 00013: val_loss improved from 0.00133 to 0.00132, saving model to weights/modelweights_Adagrad.h5
1712/1712 [==============================] - 2s 1ms/step - loss: 8.2363e-04 - acc: 0.8183 - val_loss: 0.0013 - val_acc: 0.7570
Epoch 14/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.1705e-04 - acc: 0.8060Epoch 00014: val_loss improved from 0.00132 to 0.00131, saving model to weights/modelweights_Adagrad.h5
1712/1712 [==============================] - 2s 1ms/step - loss: 8.1596e-04 - acc: 0.8072 - val_loss: 0.0013 - val_acc: 0.7570
Epoch 15/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.2026e-04 - acc: 0.8137Epoch 00015: val_loss improved from 0.00131 to 0.00131, saving model to weights/modelweights_Adagrad.h5
1712/1712 [==============================] - 2s 1ms/step - loss: 8.2113e-04 - acc: 0.8137 - val_loss: 0.0013 - val_acc: 0.7687
Epoch 16/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.2545e-04 - acc: 0.8031Epoch 00016: val_loss did not improve
1712/1712 [==============================] - 2s 970us/step - loss: 8.2521e-04 - acc: 0.8014 - val_loss: 0.0013 - val_acc: 0.7664
Epoch 17/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.0316e-04 - acc: 0.8119Epoch 00017: val_loss improved from 0.00131 to 0.00131, saving model to weights/modelweights_Adagrad.h5
1712/1712 [==============================] - 2s 1ms/step - loss: 8.0423e-04 - acc: 0.8125 - val_loss: 0.0013 - val_acc: 0.7617
Epoch 18/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.3861e-04 - acc: 0.8160Epoch 00018: val_loss did not improve
1712/1712 [==============================] - 2s 969us/step - loss: 8.3728e-04 - acc: 0.8172 - val_loss: 0.0013 - val_acc: 0.7523
Epoch 19/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.1562e-04 - acc: 0.8084Epoch 00019: val_loss did not improve
1712/1712 [==============================] - 2s 972us/step - loss: 8.1580e-04 - acc: 0.8096 - val_loss: 0.0013 - val_acc: 0.7570
Epoch 20/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.2888e-04 - acc: 0.8184Epoch 00020: val_loss did not improve
1712/1712 [==============================] - 2s 971us/step - loss: 8.2926e-04 - acc: 0.8183 - val_loss: 0.0013 - val_acc: 0.7570
Epoch 21/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.1895e-04 - acc: 0.8072Epoch 00021: val_loss did not improve
1712/1712 [==============================] - 2s 970us/step - loss: 8.1756e-04 - acc: 0.8067 - val_loss: 0.0013 - val_acc: 0.7640
Epoch 22/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.1894e-04 - acc: 0.8166Epoch 00022: val_loss did not improve
1712/1712 [==============================] - 2s 972us/step - loss: 8.1894e-04 - acc: 0.8172 - val_loss: 0.0013 - val_acc: 0.7523
Epoch 23/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.0931e-04 - acc: 0.8054Epoch 00023: val_loss did not improve
1712/1712 [==============================] - 2s 966us/step - loss: 8.0976e-04 - acc: 0.8055 - val_loss: 0.0013 - val_acc: 0.7617
Epoch 24/100
1696/1712 [============================>.] - ETA: 0s - loss: 7.8689e-04 - acc: 0.8125Epoch 00024: val_loss improved from 0.00131 to 0.00131, saving model to weights/modelweights_Adagrad.h5
1712/1712 [==============================] - 2s 1ms/step - loss: 7.8630e-04 - acc: 0.8119 - val_loss: 0.0013 - val_acc: 0.7593
Epoch 25/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.0506e-04 - acc: 0.8272Epoch 00025: val_loss improved from 0.00131 to 0.00131, saving model to weights/modelweights_Adagrad.h5
1712/1712 [==============================] - 2s 1ms/step - loss: 8.0570e-04 - acc: 0.8265 - val_loss: 0.0013 - val_acc: 0.7547
Epoch 26/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.1315e-04 - acc: 0.8101Epoch 00026: val_loss did not improve
1712/1712 [==============================] - 2s 969us/step - loss: 8.1561e-04 - acc: 0.8102 - val_loss: 0.0013 - val_acc: 0.7593
Epoch 27/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.3188e-04 - acc: 0.8048Epoch 00027: val_loss did not improve
1712/1712 [==============================] - 2s 967us/step - loss: 8.3085e-04 - acc: 0.8061 - val_loss: 0.0013 - val_acc: 0.7453
Epoch 28/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.1922e-04 - acc: 0.8166Epoch 00028: val_loss did not improve
1712/1712 [==============================] - 2s 969us/step - loss: 8.1746e-04 - acc: 0.8178 - val_loss: 0.0013 - val_acc: 0.7523
Epoch 29/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.0596e-04 - acc: 0.8096Epoch 00029: val_loss did not improve
1712/1712 [==============================] - 2s 967us/step - loss: 8.0594e-04 - acc: 0.8096 - val_loss: 0.0013 - val_acc: 0.7617
Epoch 30/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.0051e-04 - acc: 0.8107Epoch 00030: val_loss did not improve
1712/1712 [==============================] - 2s 969us/step - loss: 7.9875e-04 - acc: 0.8102 - val_loss: 0.0013 - val_acc: 0.7570
Epoch 31/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.1548e-04 - acc: 0.8172Epoch 00031: val_loss did not improve
1712/1712 [==============================] - 2s 967us/step - loss: 8.1578e-04 - acc: 0.8178 - val_loss: 0.0013 - val_acc: 0.7547
Epoch 32/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.0535e-04 - acc: 0.8172Epoch 00032: val_loss did not improve
1712/1712 [==============================] - 2s 973us/step - loss: 8.0563e-04 - acc: 0.8160 - val_loss: 0.0013 - val_acc: 0.7593
Epoch 33/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.1645e-04 - acc: 0.8149Epoch 00033: val_loss did not improve
1712/1712 [==============================] - 2s 971us/step - loss: 8.1445e-04 - acc: 0.8148 - val_loss: 0.0013 - val_acc: 0.7617
Epoch 34/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.1679e-04 - acc: 0.8101Epoch 00034: val_loss did not improve
1712/1712 [==============================] - 2s 969us/step - loss: 8.1806e-04 - acc: 0.8102 - val_loss: 0.0013 - val_acc: 0.7570
Epoch 35/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.0971e-04 - acc: 0.8154Epoch 00035: val_loss did not improve
1712/1712 [==============================] - 2s 965us/step - loss: 8.0981e-04 - acc: 0.8160 - val_loss: 0.0013 - val_acc: 0.7570
Epoch 36/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.1092e-04 - acc: 0.8137Epoch 00036: val_loss did not improve
1712/1712 [==============================] - 2s 971us/step - loss: 8.1049e-04 - acc: 0.8137 - val_loss: 0.0013 - val_acc: 0.7593
Epoch 37/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.0971e-04 - acc: 0.8208Epoch 00037: val_loss did not improve
1712/1712 [==============================] - 2s 972us/step - loss: 8.1215e-04 - acc: 0.8224 - val_loss: 0.0013 - val_acc: 0.7593
Epoch 38/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.0924e-04 - acc: 0.8190Epoch 00038: val_loss did not improve
1712/1712 [==============================] - 2s 974us/step - loss: 8.0785e-04 - acc: 0.8201 - val_loss: 0.0013 - val_acc: 0.7640
Epoch 39/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.0457e-04 - acc: 0.8261Epoch 00039: val_loss did not improve
1712/1712 [==============================] - 2s 977us/step - loss: 8.0402e-04 - acc: 0.8271 - val_loss: 0.0013 - val_acc: 0.7593
Epoch 40/100
1696/1712 [============================>.] - ETA: 0s - loss: 7.9729e-04 - acc: 0.8272Epoch 00040: val_loss did not improve
1712/1712 [==============================] - 2s 972us/step - loss: 7.9711e-04 - acc: 0.8265 - val_loss: 0.0013 - val_acc: 0.7617
Epoch 41/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.1172e-04 - acc: 0.8048- ETA: 1s - loss: 8Epoch 00041: val_loss did not improve
1712/1712 [==============================] - 2s 970us/step - loss: 8.1048e-04 - acc: 0.8043 - val_loss: 0.0013 - val_acc: 0.7570
Epoch 42/100
1696/1712 [============================>.] - ETA: 0s - loss: 7.8635e-04 - acc: 0.8113Epoch 00042: val_loss did not improve
1712/1712 [==============================] - 2s 968us/step - loss: 7.8685e-04 - acc: 0.8102 - val_loss: 0.0013 - val_acc: 0.7640
Epoch 43/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.0791e-04 - acc: 0.8225Epoch 00043: val_loss did not improve
1712/1712 [==============================] - 2s 967us/step - loss: 8.0721e-04 - acc: 0.8218 - val_loss: 0.0013 - val_acc: 0.7570
Epoch 44/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.0701e-04 - acc: 0.8166Epoch 00044: val_loss did not improve
1712/1712 [==============================] - 2s 967us/step - loss: 8.0505e-04 - acc: 0.8172 - val_loss: 0.0013 - val_acc: 0.7453
Epoch 45/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.1587e-04 - acc: 0.8149Epoch 00045: val_loss did not improve
1712/1712 [==============================] - 2s 968us/step - loss: 8.1437e-04 - acc: 0.8154 - val_loss: 0.0013 - val_acc: 0.7500
Epoch 46/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.0314e-04 - acc: 0.8290Epoch 00046: val_loss did not improve
1712/1712 [==============================] - 2s 966us/step - loss: 8.0321e-04 - acc: 0.8283 - val_loss: 0.0013 - val_acc: 0.7664
Epoch 47/100
1696/1712 [============================>.] - ETA: 0s - loss: 7.8911e-04 - acc: 0.8078Epoch 00047: val_loss did not improve
1712/1712 [==============================] - 2s 972us/step - loss: 7.8757e-04 - acc: 0.8090 - val_loss: 0.0013 - val_acc: 0.7570
Epoch 48/100
1696/1712 [============================>.] - ETA: 0s - loss: 7.9546e-04 - acc: 0.8196Epoch 00048: val_loss did not improve
1712/1712 [==============================] - 2s 971us/step - loss: 7.9461e-04 - acc: 0.8183 - val_loss: 0.0013 - val_acc: 0.7547
Epoch 49/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.0049e-04 - acc: 0.8178Epoch 00049: val_loss did not improve
1712/1712 [==============================] - 2s 970us/step - loss: 7.9951e-04 - acc: 0.8178 - val_loss: 0.0013 - val_acc: 0.7640
Epoch 50/100
1696/1712 [============================>.] - ETA: 0s - loss: 7.9202e-04 - acc: 0.8019Epoch 00050: val_loss did not improve
1712/1712 [==============================] - 2s 970us/step - loss: 7.9234e-04 - acc: 0.8026 - val_loss: 0.0013 - val_acc: 0.7500
Epoch 51/100
1696/1712 [============================>.] - ETA: 0s - loss: 7.9028e-04 - acc: 0.8113Epoch 00051: val_loss did not improve
1712/1712 [==============================] - 2s 971us/step - loss: 7.8968e-04 - acc: 0.8113 - val_loss: 0.0013 - val_acc: 0.7640
Epoch 52/100
1696/1712 [============================>.] - ETA: 0s - loss: 7.9036e-04 - acc: 0.8231Epoch 00052: val_loss did not improve
1712/1712 [==============================] - 2s 968us/step - loss: 7.8915e-04 - acc: 0.8224 - val_loss: 0.0013 - val_acc: 0.7593
Epoch 53/100
1696/1712 [============================>.] - ETA: 0s - loss: 7.8425e-04 - acc: 0.8096Epoch 00053: val_loss did not improve
1712/1712 [==============================] - 2s 967us/step - loss: 7.8459e-04 - acc: 0.8107 - val_loss: 0.0013 - val_acc: 0.7617
Epoch 54/100
1696/1712 [============================>.] - ETA: 0s - loss: 7.9142e-04 - acc: 0.8184Epoch 00054: val_loss did not improve
1712/1712 [==============================] - 2s 972us/step - loss: 7.9187e-04 - acc: 0.8189 - val_loss: 0.0013 - val_acc: 0.7617
Epoch 55/100
1696/1712 [============================>.] - ETA: 0s - loss: 7.9984e-04 - acc: 0.8054Epoch 00055: val_loss did not improve
1712/1712 [==============================] - 2s 969us/step - loss: 7.9861e-04 - acc: 0.8049 - val_loss: 0.0013 - val_acc: 0.7570
Epoch 56/100
1696/1712 [============================>.] - ETA: 0s - loss: 7.9967e-04 - acc: 0.8042Epoch 00056: val_loss did not improve
1712/1712 [==============================] - 2s 968us/step - loss: 7.9926e-04 - acc: 0.8049 - val_loss: 0.0013 - val_acc: 0.7593
Epoch 57/100
1696/1712 [============================>.] - ETA: 0s - loss: 7.8782e-04 - acc: 0.8113Epoch 00057: val_loss did not improve
1712/1712 [==============================] - 2s 967us/step - loss: 7.8753e-04 - acc: 0.8131 - val_loss: 0.0013 - val_acc: 0.7617
Epoch 58/100
1696/1712 [============================>.] - ETA: 0s - loss: 7.8819e-04 - acc: 0.8208Epoch 00058: val_loss did not improve
1712/1712 [==============================] - 2s 966us/step - loss: 7.9000e-04 - acc: 0.8207 - val_loss: 0.0013 - val_acc: 0.7547
Epoch 59/100
1696/1712 [============================>.] - ETA: 0s - loss: 7.9044e-04 - acc: 0.8131Epoch 00059: val_loss did not improve
1712/1712 [==============================] - 2s 975us/step - loss: 7.9003e-04 - acc: 0.8125 - val_loss: 0.0013 - val_acc: 0.7570
Epoch 60/100
1696/1712 [============================>.] - ETA: 0s - loss: 7.9729e-04 - acc: 0.8172Epoch 00060: val_loss did not improve
1712/1712 [==============================] - 2s 973us/step - loss: 7.9467e-04 - acc: 0.8178 - val_loss: 0.0013 - val_acc: 0.7664
Epoch 61/100
1696/1712 [============================>.] - ETA: 0s - loss: 7.9470e-04 - acc: 0.8119Epoch 00061: val_loss did not improve
1712/1712 [==============================] - 2s 971us/step - loss: 7.9477e-04 - acc: 0.8113 - val_loss: 0.0013 - val_acc: 0.7593
Epoch 62/100
1696/1712 [============================>.] - ETA: 0s - loss: 7.9078e-04 - acc: 0.8160Epoch 00062: val_loss did not improve
1712/1712 [==============================] - 2s 969us/step - loss: 7.9152e-04 - acc: 0.8160 - val_loss: 0.0013 - val_acc: 0.7664
Epoch 63/100
1696/1712 [============================>.] - ETA: 0s - loss: 7.9405e-04 - acc: 0.8154Epoch 00063: val_loss did not improve
1712/1712 [==============================] - 2s 971us/step - loss: 7.9520e-04 - acc: 0.8154 - val_loss: 0.0013 - val_acc: 0.7687
Epoch 64/100
1696/1712 [============================>.] - ETA: 0s - loss: 7.9733e-04 - acc: 0.8202Epoch 00064: val_loss did not improve
1712/1712 [==============================] - 2s 968us/step - loss: 7.9738e-04 - acc: 0.8201 - val_loss: 0.0013 - val_acc: 0.7640
Epoch 65/100
1696/1712 [============================>.] - ETA: 0s - loss: 7.8774e-04 - acc: 0.8149Epoch 00065: val_loss did not improve
1712/1712 [==============================] - 2s 970us/step - loss: 7.8626e-04 - acc: 0.8143 - val_loss: 0.0013 - val_acc: 0.7570
Epoch 66/100
1696/1712 [============================>.] - ETA: 0s - loss: 7.9863e-04 - acc: 0.8261Epoch 00066: val_loss did not improve
1712/1712 [==============================] - 2s 965us/step - loss: 7.9814e-04 - acc: 0.8254 - val_loss: 0.0013 - val_acc: 0.7570
Epoch 67/100
1696/1712 [============================>.] - ETA: 0s - loss: 7.9401e-04 - acc: 0.8202Epoch 00067: val_loss did not improve
1712/1712 [==============================] - 2s 968us/step - loss: 8.0076e-04 - acc: 0.8207 - val_loss: 0.0013 - val_acc: 0.7640
Epoch 68/100
1696/1712 [============================>.] - ETA: 0s - loss: 7.9877e-04 - acc: 0.8131Epoch 00068: val_loss did not improve
1712/1712 [==============================] - 2s 973us/step - loss: 7.9776e-04 - acc: 0.8137 - val_loss: 0.0013 - val_acc: 0.7617
Epoch 69/100
1696/1712 [============================>.] - ETA: 0s - loss: 7.7567e-04 - acc: 0.8149Epoch 00069: val_loss did not improve
1712/1712 [==============================] - 2s 972us/step - loss: 7.7722e-04 - acc: 0.8154 - val_loss: 0.0013 - val_acc: 0.7547
Epoch 70/100
1696/1712 [============================>.] - ETA: 0s - loss: 7.9103e-04 - acc: 0.8166Epoch 00070: val_loss did not improve
1712/1712 [==============================] - 2s 969us/step - loss: 7.9007e-04 - acc: 0.8178 - val_loss: 0.0013 - val_acc: 0.7617
Epoch 71/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.0449e-04 - acc: 0.8119Epoch 00071: val_loss did not improve
1712/1712 [==============================] - 2s 966us/step - loss: 8.0664e-04 - acc: 0.8107 - val_loss: 0.0013 - val_acc: 0.7593
Epoch 72/100
1696/1712 [============================>.] - ETA: 0s - loss: 7.9475e-04 - acc: 0.8060Epoch 00072: val_loss did not improve
1712/1712 [==============================] - 2s 969us/step - loss: 7.9234e-04 - acc: 0.8061 - val_loss: 0.0013 - val_acc: 0.7617
Epoch 73/100
1696/1712 [============================>.] - ETA: 0s - loss: 7.7826e-04 - acc: 0.8202Epoch 00073: val_loss did not improve
1712/1712 [==============================] - 2s 969us/step - loss: 7.8096e-04 - acc: 0.8183 - val_loss: 0.0013 - val_acc: 0.7593
Epoch 74/100
1696/1712 [============================>.] - ETA: 0s - loss: 7.9219e-04 - acc: 0.8072Epoch 00074: val_loss did not improve
1712/1712 [==============================] - 2s 969us/step - loss: 7.9098e-04 - acc: 0.8072 - val_loss: 0.0013 - val_acc: 0.7687
Epoch 75/100
1696/1712 [============================>.] - ETA: 0s - loss: 7.9031e-04 - acc: 0.8178Epoch 00075: val_loss did not improve
1712/1712 [==============================] - 2s 970us/step - loss: 7.8958e-04 - acc: 0.8178 - val_loss: 0.0013 - val_acc: 0.7664
Epoch 76/100
1696/1712 [============================>.] - ETA: 0s - loss: 7.9075e-04 - acc: 0.7995Epoch 00076: val_loss did not improve
1712/1712 [==============================] - 2s 965us/step - loss: 7.8843e-04 - acc: 0.7996 - val_loss: 0.0013 - val_acc: 0.7687
Epoch 77/100
1696/1712 [============================>.] - ETA: 0s - loss: 7.9245e-04 - acc: 0.8125Epoch 00077: val_loss did not improve
1712/1712 [==============================] - 2s 965us/step - loss: 7.9170e-04 - acc: 0.8125 - val_loss: 0.0013 - val_acc: 0.7617
Epoch 78/100
1696/1712 [============================>.] - ETA: 0s - loss: 7.9834e-04 - acc: 0.8154Epoch 00078: val_loss did not improve
1712/1712 [==============================] - 2s 969us/step - loss: 7.9991e-04 - acc: 0.8166 - val_loss: 0.0013 - val_acc: 0.7570
Epoch 79/100
1696/1712 [============================>.] - ETA: 0s - loss: 7.9157e-04 - acc: 0.8113Epoch 00079: val_loss did not improve
1712/1712 [==============================] - 2s 971us/step - loss: 7.9136e-04 - acc: 0.8102 - val_loss: 0.0013 - val_acc: 0.7593
Epoch 80/100
1696/1712 [============================>.] - ETA: 0s - loss: 7.9202e-04 - acc: 0.8096Epoch 00080: val_loss did not improve
1712/1712 [==============================] - 2s 967us/step - loss: 7.9239e-04 - acc: 0.8102 - val_loss: 0.0013 - val_acc: 0.7593
Epoch 81/100
1696/1712 [============================>.] - ETA: 0s - loss: 7.8000e-04 - acc: 0.8160Epoch 00081: val_loss did not improve
1712/1712 [==============================] - 2s 965us/step - loss: 7.7893e-04 - acc: 0.8166 - val_loss: 0.0013 - val_acc: 0.7547
Epoch 82/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.0632e-04 - acc: 0.8090Epoch 00082: val_loss did not improve
1712/1712 [==============================] - 2s 967us/step - loss: 8.0390e-04 - acc: 0.8096 - val_loss: 0.0013 - val_acc: 0.7593
Epoch 83/100
1696/1712 [============================>.] - ETA: 0s - loss: 7.9271e-04 - acc: 0.8060Epoch 00083: val_loss did not improve
1712/1712 [==============================] - 2s 969us/step - loss: 7.9387e-04 - acc: 0.8061 - val_loss: 0.0013 - val_acc: 0.7570
Epoch 84/100
1696/1712 [============================>.] - ETA: 0s - loss: 7.9805e-04 - acc: 0.8119Epoch 00084: val_loss did not improve
1712/1712 [==============================] - 2s 968us/step - loss: 8.0064e-04 - acc: 0.8113 - val_loss: 0.0013 - val_acc: 0.7570
Epoch 85/100
1696/1712 [============================>.] - ETA: 0s - loss: 7.9096e-04 - acc: 0.8066Epoch 00085: val_loss did not improve
1712/1712 [==============================] - 2s 965us/step - loss: 7.8925e-04 - acc: 0.8084 - val_loss: 0.0013 - val_acc: 0.7617
Epoch 86/100
1696/1712 [============================>.] - ETA: 0s - loss: 7.8134e-04 - acc: 0.8137Epoch 00086: val_loss did not improve
1712/1712 [==============================] - 2s 963us/step - loss: 7.8260e-04 - acc: 0.8131 - val_loss: 0.0013 - val_acc: 0.7664
Epoch 87/100
1696/1712 [============================>.] - ETA: 0s - loss: 7.8650e-04 - acc: 0.8119Epoch 00087: val_loss did not improve
1712/1712 [==============================] - 2s 970us/step - loss: 7.8842e-04 - acc: 0.8113 - val_loss: 0.0013 - val_acc: 0.7640
Epoch 88/100
1696/1712 [============================>.] - ETA: 0s - loss: 7.7746e-04 - acc: 0.8107Epoch 00088: val_loss did not improve
1712/1712 [==============================] - 2s 968us/step - loss: 7.7844e-04 - acc: 0.8113 - val_loss: 0.0013 - val_acc: 0.7617
Epoch 89/100
1696/1712 [============================>.] - ETA: 0s - loss: 7.8230e-04 - acc: 0.8160Epoch 00089: val_loss did not improve
1712/1712 [==============================] - 2s 964us/step - loss: 7.8218e-04 - acc: 0.8160 - val_loss: 0.0013 - val_acc: 0.7617
Epoch 90/100
1696/1712 [============================>.] - ETA: 0s - loss: 7.9473e-04 - acc: 0.8225Epoch 00090: val_loss did not improve
1712/1712 [==============================] - 2s 968us/step - loss: 8.0199e-04 - acc: 0.8207 - val_loss: 0.0013 - val_acc: 0.7664
Epoch 91/100
1696/1712 [============================>.] - ETA: 0s - loss: 7.8130e-04 - acc: 0.8090Epoch 00091: val_loss did not improve
1712/1712 [==============================] - 2s 968us/step - loss: 7.8318e-04 - acc: 0.8072 - val_loss: 0.0013 - val_acc: 0.7640
Epoch 92/100
1696/1712 [============================>.] - ETA: 0s - loss: 7.7613e-04 - acc: 0.8154Epoch 00092: val_loss did not improve
1712/1712 [==============================] - 2s 964us/step - loss: 7.7429e-04 - acc: 0.8166 - val_loss: 0.0013 - val_acc: 0.7547
Epoch 93/100
1696/1712 [============================>.] - ETA: 0s - loss: 7.7289e-04 - acc: 0.8178Epoch 00093: val_loss did not improve
1712/1712 [==============================] - 2s 968us/step - loss: 7.7251e-04 - acc: 0.8178 - val_loss: 0.0013 - val_acc: 0.7593
Epoch 94/100
1696/1712 [============================>.] - ETA: 0s - loss: 7.8815e-04 - acc: 0.8290Epoch 00094: val_loss did not improve
1712/1712 [==============================] - 2s 970us/step - loss: 7.8845e-04 - acc: 0.8277 - val_loss: 0.0013 - val_acc: 0.7617
Epoch 95/100
1696/1712 [============================>.] - ETA: 0s - loss: 7.9075e-04 - acc: 0.8160Epoch 00095: val_loss did not improve
1712/1712 [==============================] - 2s 969us/step - loss: 7.8963e-04 - acc: 0.8160 - val_loss: 0.0013 - val_acc: 0.7593
Epoch 96/100
1696/1712 [============================>.] - ETA: 0s - loss: 7.7515e-04 - acc: 0.8290- ETA: 1s - loss: 7Epoch 00096: val_loss did not improve
1712/1712 [==============================] - 2s 970us/step - loss: 7.7646e-04 - acc: 0.8289 - val_loss: 0.0013 - val_acc: 0.7593
Epoch 97/100
1696/1712 [============================>.] - ETA: 0s - loss: 7.8253e-04 - acc: 0.8249Epoch 00097: val_loss did not improve
1712/1712 [==============================] - 2s 969us/step - loss: 7.8232e-04 - acc: 0.8248 - val_loss: 0.0013 - val_acc: 0.7664
Epoch 98/100
1696/1712 [============================>.] - ETA: 0s - loss: 7.7257e-04 - acc: 0.8125Epoch 00098: val_loss did not improve
1712/1712 [==============================] - 2s 967us/step - loss: 7.7354e-04 - acc: 0.8125 - val_loss: 0.0013 - val_acc: 0.7593
Epoch 99/100
1696/1712 [============================>.] - ETA: 0s - loss: 7.7873e-04 - acc: 0.8125Epoch 00099: val_loss did not improve
1712/1712 [==============================] - 2s 966us/step - loss: 7.7736e-04 - acc: 0.8119 - val_loss: 0.0013 - val_acc: 0.7617
Epoch 100/100
1696/1712 [============================>.] - ETA: 0s - loss: 7.9505e-04 - acc: 0.8066Epoch 00100: val_loss did not improve
1712/1712 [==============================] - 2s 969us/step - loss: 7.9226e-04 - acc: 0.8078 - val_loss: 0.0013 - val_acc: 0.7593


 2017-12-11 23:34:55.543738
Using optimizer:  RMSprop
Did not load old weights...
Train on 1712 samples, validate on 428 samples
Epoch 1/100
1696/1712 [============================>.] - ETA: 0s - loss: 9.0574e-04 - acc: 0.7960Epoch 00001: val_loss improved from inf to 0.00139, saving model to weights/modelweights_RMSprop.h5
1712/1712 [==============================] - 2s 1ms/step - loss: 9.0513e-04 - acc: 0.7979 - val_loss: 0.0014 - val_acc: 0.7664
Epoch 2/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.7463e-04 - acc: 0.8143Epoch 00002: val_loss did not improve
1712/1712 [==============================] - 2s 972us/step - loss: 8.7837e-04 - acc: 0.8131 - val_loss: 0.0014 - val_acc: 0.7734
Epoch 3/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.6430e-04 - acc: 0.8125Epoch 00003: val_loss improved from 0.00139 to 0.00136, saving model to weights/modelweights_RMSprop.h5
1712/1712 [==============================] - 2s 1ms/step - loss: 8.6750e-04 - acc: 0.8131 - val_loss: 0.0014 - val_acc: 0.7523
Epoch 4/100
1696/1712 [============================>.] - ETA: 0s - loss: 9.0017e-04 - acc: 0.8072Epoch 00004: val_loss did not improve
1712/1712 [==============================] - 2s 983us/step - loss: 9.0152e-04 - acc: 0.8072 - val_loss: 0.0014 - val_acc: 0.7523
Epoch 5/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.8903e-04 - acc: 0.7989Epoch 00005: val_loss improved from 0.00136 to 0.00135, saving model to weights/modelweights_RMSprop.h5
1712/1712 [==============================] - 2s 1ms/step - loss: 8.8902e-04 - acc: 0.7996 - val_loss: 0.0013 - val_acc: 0.7617
Epoch 6/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.8771e-04 - acc: 0.8096Epoch 00006: val_loss did not improve
1712/1712 [==============================] - 2s 977us/step - loss: 8.8584e-04 - acc: 0.8102 - val_loss: 0.0014 - val_acc: 0.7523
Epoch 7/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.8487e-04 - acc: 0.8019Epoch 00007: val_loss did not improve
1712/1712 [==============================] - 2s 975us/step - loss: 8.8876e-04 - acc: 0.8008 - val_loss: 0.0014 - val_acc: 0.7593
Epoch 8/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.7669e-04 - acc: 0.7960Epoch 00008: val_loss did not improve
1712/1712 [==============================] - 2s 976us/step - loss: 8.7741e-04 - acc: 0.7956 - val_loss: 0.0014 - val_acc: 0.7593
Epoch 9/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.7709e-04 - acc: 0.8025Epoch 00009: val_loss did not improve
1712/1712 [==============================] - 2s 976us/step - loss: 8.7651e-04 - acc: 0.8032 - val_loss: 0.0014 - val_acc: 0.7593
Epoch 10/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.8942e-04 - acc: 0.8007Epoch 00010: val_loss did not improve
1712/1712 [==============================] - 2s 981us/step - loss: 8.8904e-04 - acc: 0.8008 - val_loss: 0.0014 - val_acc: 0.7547
Epoch 11/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.8919e-04 - acc: 0.8048Epoch 00011: val_loss did not improve
1712/1712 [==============================] - 2s 986us/step - loss: 8.8740e-04 - acc: 0.8032 - val_loss: 0.0014 - val_acc: 0.7523
Epoch 12/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.8749e-04 - acc: 0.8031Epoch 00012: val_loss did not improve
1712/1712 [==============================] - 2s 979us/step - loss: 8.8895e-04 - acc: 0.8043 - val_loss: 0.0014 - val_acc: 0.7547
Epoch 13/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.8999e-04 - acc: 0.8072Epoch 00013: val_loss did not improve
1712/1712 [==============================] - 2s 975us/step - loss: 8.8924e-04 - acc: 0.8078 - val_loss: 0.0014 - val_acc: 0.7710
Epoch 14/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.8453e-04 - acc: 0.8119Epoch 00014: val_loss did not improve
1712/1712 [==============================] - 2s 977us/step - loss: 8.8425e-04 - acc: 0.8119 - val_loss: 0.0014 - val_acc: 0.7664
Epoch 15/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.8277e-04 - acc: 0.8001Epoch 00015: val_loss did not improve
1712/1712 [==============================] - 2s 981us/step - loss: 8.8294e-04 - acc: 0.8002 - val_loss: 0.0014 - val_acc: 0.7570
Epoch 16/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.7384e-04 - acc: 0.8137Epoch 00016: val_loss did not improve
1712/1712 [==============================] - 2s 980us/step - loss: 8.7298e-04 - acc: 0.8137 - val_loss: 0.0014 - val_acc: 0.7687
Epoch 17/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.9538e-04 - acc: 0.8084Epoch 00017: val_loss improved from 0.00135 to 0.00135, saving model to weights/modelweights_RMSprop.h5
1712/1712 [==============================] - 2s 1ms/step - loss: 8.9589e-04 - acc: 0.8078 - val_loss: 0.0013 - val_acc: 0.7687
Epoch 18/100
1696/1712 [============================>.] - ETA: 0s - loss: 9.0972e-04 - acc: 0.7936Epoch 00018: val_loss did not improve
1712/1712 [==============================] - 2s 974us/step - loss: 9.0907e-04 - acc: 0.7932 - val_loss: 0.0014 - val_acc: 0.7477
Epoch 19/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.6882e-04 - acc: 0.8013Epoch 00019: val_loss did not improve
1712/1712 [==============================] - 2s 976us/step - loss: 8.6788e-04 - acc: 0.8002 - val_loss: 0.0014 - val_acc: 0.7710
Epoch 20/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.9479e-04 - acc: 0.8090Epoch 00020: val_loss did not improve
1712/1712 [==============================] - 2s 979us/step - loss: 8.9471e-04 - acc: 0.8090 - val_loss: 0.0014 - val_acc: 0.7617
Epoch 21/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.7213e-04 - acc: 0.8042Epoch 00021: val_loss did not improve
1712/1712 [==============================] - 2s 980us/step - loss: 8.7101e-04 - acc: 0.8043 - val_loss: 0.0014 - val_acc: 0.7734
Epoch 22/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.9190e-04 - acc: 0.8054Epoch 00022: val_loss did not improve
1712/1712 [==============================] - 2s 974us/step - loss: 8.9165e-04 - acc: 0.8055 - val_loss: 0.0014 - val_acc: 0.7523
Epoch 23/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.9386e-04 - acc: 0.8013Epoch 00023: val_loss did not improve
1712/1712 [==============================] - 2s 974us/step - loss: 8.9382e-04 - acc: 0.8002 - val_loss: 0.0014 - val_acc: 0.7593
Epoch 24/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.8061e-04 - acc: 0.8054Epoch 00024: val_loss did not improve
1712/1712 [==============================] - 2s 974us/step - loss: 8.7953e-04 - acc: 0.8067 - val_loss: 0.0014 - val_acc: 0.7500
Epoch 25/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.9118e-04 - acc: 0.7972Epoch 00025: val_loss improved from 0.00135 to 0.00134, saving model to weights/modelweights_RMSprop.h5
1712/1712 [==============================] - 2s 1ms/step - loss: 8.9204e-04 - acc: 0.7979 - val_loss: 0.0013 - val_acc: 0.7687
Epoch 26/100
1696/1712 [============================>.] - ETA: 0s - loss: 9.0127e-04 - acc: 0.8096Epoch 00026: val_loss did not improve
1712/1712 [==============================] - 2s 979us/step - loss: 9.0089e-04 - acc: 0.8096 - val_loss: 0.0014 - val_acc: 0.7500
Epoch 27/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.7121e-04 - acc: 0.8231Epoch 00027: val_loss did not improve
1712/1712 [==============================] - 2s 977us/step - loss: 8.7644e-04 - acc: 0.8207 - val_loss: 0.0013 - val_acc: 0.7547
Epoch 28/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.8906e-04 - acc: 0.7983Epoch 00028: val_loss did not improve
1712/1712 [==============================] - 2s 970us/step - loss: 8.9375e-04 - acc: 0.7979 - val_loss: 0.0014 - val_acc: 0.7664
Epoch 29/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.8717e-04 - acc: 0.8090Epoch 00029: val_loss improved from 0.00134 to 0.00133, saving model to weights/modelweights_RMSprop.h5
1712/1712 [==============================] - 2s 1ms/step - loss: 8.8517e-04 - acc: 0.8090 - val_loss: 0.0013 - val_acc: 0.7570
Epoch 30/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.8946e-04 - acc: 0.8019Epoch 00030: val_loss improved from 0.00133 to 0.00133, saving model to weights/modelweights_RMSprop.h5
1712/1712 [==============================] - 2s 1ms/step - loss: 8.8685e-04 - acc: 0.8032 - val_loss: 0.0013 - val_acc: 0.7593
Epoch 31/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.6891e-04 - acc: 0.8025Epoch 00031: val_loss did not improve
1712/1712 [==============================] - 2s 968us/step - loss: 8.7975e-04 - acc: 0.8008 - val_loss: 0.0014 - val_acc: 0.7617
Epoch 32/100
1696/1712 [============================>.] - ETA: 0s - loss: 9.0250e-04 - acc: 0.8090Epoch 00032: val_loss did not improve
1712/1712 [==============================] - 2s 970us/step - loss: 9.0218e-04 - acc: 0.8090 - val_loss: 0.0014 - val_acc: 0.7523
Epoch 33/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.6694e-04 - acc: 0.8066Epoch 00033: val_loss did not improve
1712/1712 [==============================] - 2s 970us/step - loss: 8.6655e-04 - acc: 0.8072 - val_loss: 0.0014 - val_acc: 0.7500
Epoch 34/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.7799e-04 - acc: 0.7948Epoch 00034: val_loss did not improve
1712/1712 [==============================] - 2s 965us/step - loss: 8.7848e-04 - acc: 0.7944 - val_loss: 0.0014 - val_acc: 0.7477
Epoch 35/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.7886e-04 - acc: 0.8037Epoch 00035: val_loss did not improve
1712/1712 [==============================] - 2s 970us/step - loss: 8.7788e-04 - acc: 0.8020 - val_loss: 0.0014 - val_acc: 0.7477
Epoch 36/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.6313e-04 - acc: 0.8143Epoch 00036: val_loss did not improve
1712/1712 [==============================] - 2s 969us/step - loss: 8.6234e-04 - acc: 0.8154 - val_loss: 0.0014 - val_acc: 0.7687
Epoch 37/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.7542e-04 - acc: 0.8143Epoch 00037: val_loss did not improve
1712/1712 [==============================] - 2s 971us/step - loss: 8.7548e-04 - acc: 0.8143 - val_loss: 0.0014 - val_acc: 0.7804
Epoch 38/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.8799e-04 - acc: 0.7954Epoch 00038: val_loss did not improve
1712/1712 [==============================] - 2s 972us/step - loss: 8.8689e-04 - acc: 0.7956 - val_loss: 0.0013 - val_acc: 0.7664
Epoch 39/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.8327e-04 - acc: 0.8178Epoch 00039: val_loss did not improve
1712/1712 [==============================] - 2s 981us/step - loss: 8.8648e-04 - acc: 0.8166 - val_loss: 0.0014 - val_acc: 0.7570
Epoch 40/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.8212e-04 - acc: 0.8107Epoch 00040: val_loss did not improve
1712/1712 [==============================] - 2s 975us/step - loss: 8.8164e-04 - acc: 0.8113 - val_loss: 0.0013 - val_acc: 0.7850
Epoch 41/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.8642e-04 - acc: 0.8290Epoch 00041: val_loss did not improve
1712/1712 [==============================] - 2s 974us/step - loss: 8.8604e-04 - acc: 0.8277 - val_loss: 0.0014 - val_acc: 0.7710
Epoch 42/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.8612e-04 - acc: 0.8160Epoch 00042: val_loss did not improve
1712/1712 [==============================] - 2s 975us/step - loss: 8.8660e-04 - acc: 0.8172 - val_loss: 0.0014 - val_acc: 0.7617
Epoch 43/100
1696/1712 [============================>.] - ETA: 0s - loss: 9.0962e-04 - acc: 0.8066Epoch 00043: val_loss did not improve
1712/1712 [==============================] - 2s 976us/step - loss: 9.0858e-04 - acc: 0.8055 - val_loss: 0.0014 - val_acc: 0.7547
Epoch 44/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.9245e-04 - acc: 0.7930Epoch 00044: val_loss did not improve
1712/1712 [==============================] - 2s 976us/step - loss: 8.9158e-04 - acc: 0.7926 - val_loss: 0.0014 - val_acc: 0.7780
Epoch 45/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.9688e-04 - acc: 0.8013Epoch 00045: val_loss did not improve
1712/1712 [==============================] - 2s 975us/step - loss: 8.9721e-04 - acc: 0.8026 - val_loss: 0.0014 - val_acc: 0.7523
Epoch 46/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.6668e-04 - acc: 0.8137Epoch 00046: val_loss did not improve
1712/1712 [==============================] - 2s 973us/step - loss: 8.6840e-04 - acc: 0.8137 - val_loss: 0.0014 - val_acc: 0.7664
Epoch 47/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.7753e-04 - acc: 0.8060Epoch 00047: val_loss did not improve
1712/1712 [==============================] - 2s 976us/step - loss: 8.7599e-04 - acc: 0.8061 - val_loss: 0.0014 - val_acc: 0.7710
Epoch 48/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.8686e-04 - acc: 0.8113Epoch 00048: val_loss did not improve
1712/1712 [==============================] - 2s 974us/step - loss: 8.8428e-04 - acc: 0.8131 - val_loss: 0.0013 - val_acc: 0.7570
Epoch 49/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.7773e-04 - acc: 0.8048Epoch 00049: val_loss did not improve
1712/1712 [==============================] - 2s 974us/step - loss: 8.7800e-04 - acc: 0.8043 - val_loss: 0.0014 - val_acc: 0.7617
Epoch 50/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.8676e-04 - acc: 0.8096Epoch 00050: val_loss did not improve
1712/1712 [==============================] - 2s 973us/step - loss: 8.8715e-04 - acc: 0.8084 - val_loss: 0.0014 - val_acc: 0.7664
Epoch 51/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.8082e-04 - acc: 0.8137Epoch 00051: val_loss improved from 0.00133 to 0.00132, saving model to weights/modelweights_RMSprop.h5
1712/1712 [==============================] - 2s 1ms/step - loss: 8.8384e-04 - acc: 0.8143 - val_loss: 0.0013 - val_acc: 0.7710
Epoch 52/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.7872e-04 - acc: 0.8172Epoch 00052: val_loss did not improve
1712/1712 [==============================] - 2s 973us/step - loss: 8.7780e-04 - acc: 0.8172 - val_loss: 0.0013 - val_acc: 0.7640
Epoch 53/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.7948e-04 - acc: 0.8037Epoch 00053: val_loss did not improve
1712/1712 [==============================] - 2s 976us/step - loss: 8.7828e-04 - acc: 0.8043 - val_loss: 0.0014 - val_acc: 0.7547
Epoch 54/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.7277e-04 - acc: 0.8125Epoch 00054: val_loss did not improve
1712/1712 [==============================] - 2s 977us/step - loss: 8.7130e-04 - acc: 0.8131 - val_loss: 0.0013 - val_acc: 0.7523
Epoch 55/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.8339e-04 - acc: 0.8113Epoch 00055: val_loss did not improve
1712/1712 [==============================] - 2s 971us/step - loss: 8.8499e-04 - acc: 0.8102 - val_loss: 0.0014 - val_acc: 0.7617
Epoch 56/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.8761e-04 - acc: 0.7978Epoch 00056: val_loss did not improve
1712/1712 [==============================] - 2s 977us/step - loss: 8.8976e-04 - acc: 0.7973 - val_loss: 0.0014 - val_acc: 0.7804
Epoch 57/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.7483e-04 - acc: 0.8143Epoch 00057: val_loss did not improve
1712/1712 [==============================] - 2s 979us/step - loss: 8.7361e-04 - acc: 0.8143 - val_loss: 0.0013 - val_acc: 0.7710
Epoch 58/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.5795e-04 - acc: 0.8042Epoch 00058: val_loss did not improve
1712/1712 [==============================] - 2s 978us/step - loss: 8.5789e-04 - acc: 0.8026 - val_loss: 0.0014 - val_acc: 0.7477
Epoch 59/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.7138e-04 - acc: 0.7930Epoch 00059: val_loss did not improve
1712/1712 [==============================] - 2s 977us/step - loss: 8.7002e-04 - acc: 0.7926 - val_loss: 0.0014 - val_acc: 0.7570
Epoch 60/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.7916e-04 - acc: 0.8119Epoch 00060: val_loss did not improve
1712/1712 [==============================] - 2s 985us/step - loss: 8.8001e-04 - acc: 0.8107 - val_loss: 0.0014 - val_acc: 0.7593
Epoch 61/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.7679e-04 - acc: 0.8166Epoch 00061: val_loss did not improve
1712/1712 [==============================] - 2s 980us/step - loss: 8.7626e-04 - acc: 0.8166 - val_loss: 0.0014 - val_acc: 0.7617
Epoch 62/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.8396e-04 - acc: 0.7989Epoch 00062: val_loss did not improve
1712/1712 [==============================] - 2s 979us/step - loss: 8.8221e-04 - acc: 0.7985 - val_loss: 0.0014 - val_acc: 0.7757
Epoch 63/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.6262e-04 - acc: 0.8013Epoch 00063: val_loss did not improve
1712/1712 [==============================] - 2s 984us/step - loss: 8.6009e-04 - acc: 0.8002 - val_loss: 0.0014 - val_acc: 0.7640
Epoch 64/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.6222e-04 - acc: 0.8137Epoch 00064: val_loss did not improve
1712/1712 [==============================] - 2s 981us/step - loss: 8.6138e-04 - acc: 0.8148 - val_loss: 0.0014 - val_acc: 0.7570
Epoch 65/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.7468e-04 - acc: 0.8119Epoch 00065: val_loss did not improve
1712/1712 [==============================] - 2s 980us/step - loss: 8.7471e-04 - acc: 0.8125 - val_loss: 0.0014 - val_acc: 0.7710
Epoch 66/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.6923e-04 - acc: 0.8119Epoch 00066: val_loss did not improve
1712/1712 [==============================] - 2s 984us/step - loss: 8.7095e-04 - acc: 0.8107 - val_loss: 0.0014 - val_acc: 0.7827
Epoch 67/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.6688e-04 - acc: 0.8090Epoch 00067: val_loss did not improve
1712/1712 [==============================] - 2s 969us/step - loss: 8.6813e-04 - acc: 0.8096 - val_loss: 0.0014 - val_acc: 0.7804
Epoch 68/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.6223e-04 - acc: 0.8107Epoch 00068: val_loss did not improve
1712/1712 [==============================] - 2s 964us/step - loss: 8.6865e-04 - acc: 0.8096 - val_loss: 0.0014 - val_acc: 0.7617
Epoch 69/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.6856e-04 - acc: 0.8160Epoch 00069: val_loss did not improve
1712/1712 [==============================] - 2s 972us/step - loss: 8.6696e-04 - acc: 0.8160 - val_loss: 0.0013 - val_acc: 0.7523
Epoch 70/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.7815e-04 - acc: 0.8137Epoch 00070: val_loss did not improve
1712/1712 [==============================] - 2s 977us/step - loss: 8.7976e-04 - acc: 0.8137 - val_loss: 0.0014 - val_acc: 0.7710
Epoch 71/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.6882e-04 - acc: 0.8190Epoch 00071: val_loss did not improve
1712/1712 [==============================] - 2s 972us/step - loss: 8.7071e-04 - acc: 0.8166 - val_loss: 0.0014 - val_acc: 0.7664
Epoch 72/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.7740e-04 - acc: 0.7948Epoch 00072: val_loss did not improve
1712/1712 [==============================] - 2s 971us/step - loss: 8.7639e-04 - acc: 0.7967 - val_loss: 0.0014 - val_acc: 0.7710
Epoch 73/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.7085e-04 - acc: 0.8166Epoch 00073: val_loss did not improve
1712/1712 [==============================] - 2s 973us/step - loss: 8.6980e-04 - acc: 0.8172 - val_loss: 0.0013 - val_acc: 0.7664
Epoch 74/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.7769e-04 - acc: 0.7942Epoch 00074: val_loss did not improve
1712/1712 [==============================] - 2s 977us/step - loss: 8.7684e-04 - acc: 0.7944 - val_loss: 0.0013 - val_acc: 0.7710
Epoch 75/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.7841e-04 - acc: 0.8078Epoch 00075: val_loss did not improve
1712/1712 [==============================] - 2s 978us/step - loss: 8.7669e-04 - acc: 0.8090 - val_loss: 0.0014 - val_acc: 0.7687
Epoch 76/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.4981e-04 - acc: 0.8025Epoch 00076: val_loss did not improve
1712/1712 [==============================] - 2s 978us/step - loss: 8.5114e-04 - acc: 0.8037 - val_loss: 0.0014 - val_acc: 0.7593
Epoch 77/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.7183e-04 - acc: 0.8113Epoch 00077: val_loss did not improve
1712/1712 [==============================] - 2s 976us/step - loss: 8.7084e-04 - acc: 0.8113 - val_loss: 0.0014 - val_acc: 0.7710
Epoch 78/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.6530e-04 - acc: 0.8042Epoch 00078: val_loss did not improve
1712/1712 [==============================] - 2s 976us/step - loss: 8.6388e-04 - acc: 0.8043 - val_loss: 0.0014 - val_acc: 0.7757
Epoch 79/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.8066e-04 - acc: 0.8166Epoch 00079: val_loss did not improve
1712/1712 [==============================] - 2s 971us/step - loss: 8.8006e-04 - acc: 0.8160 - val_loss: 0.0013 - val_acc: 0.7734
Epoch 80/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.9296e-04 - acc: 0.8096Epoch 00080: val_loss did not improve
1712/1712 [==============================] - 2s 975us/step - loss: 8.9120e-04 - acc: 0.8084 - val_loss: 0.0014 - val_acc: 0.7453
Epoch 81/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.7325e-04 - acc: 0.8113Epoch 00081: val_loss did not improve
1712/1712 [==============================] - 2s 975us/step - loss: 8.7053e-04 - acc: 0.8113 - val_loss: 0.0013 - val_acc: 0.7617
Epoch 82/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.8682e-04 - acc: 0.8084Epoch 00082: val_loss did not improve
1712/1712 [==============================] - 2s 983us/step - loss: 8.8897e-04 - acc: 0.8090 - val_loss: 0.0014 - val_acc: 0.7687
Epoch 83/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.7642e-04 - acc: 0.8284Epoch 00083: val_loss did not improve
1712/1712 [==============================] - 2s 974us/step - loss: 8.7472e-04 - acc: 0.8277 - val_loss: 0.0013 - val_acc: 0.7593
Epoch 84/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.8362e-04 - acc: 0.7995Epoch 00084: val_loss did not improve
1712/1712 [==============================] - 2s 971us/step - loss: 8.8487e-04 - acc: 0.8014 - val_loss: 0.0014 - val_acc: 0.7640
Epoch 85/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.6325e-04 - acc: 0.8048Epoch 00085: val_loss did not improve
1712/1712 [==============================] - 2s 972us/step - loss: 8.6712e-04 - acc: 0.8061 - val_loss: 0.0014 - val_acc: 0.7710
Epoch 86/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.6806e-04 - acc: 0.7972Epoch 00086: val_loss did not improve
1712/1712 [==============================] - 2s 971us/step - loss: 8.7267e-04 - acc: 0.7973 - val_loss: 0.0014 - val_acc: 0.7617
Epoch 87/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.8344e-04 - acc: 0.8001Epoch 00087: val_loss did not improve
1712/1712 [==============================] - 2s 976us/step - loss: 8.8290e-04 - acc: 0.8002 - val_loss: 0.0014 - val_acc: 0.7617
Epoch 88/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.6818e-04 - acc: 0.8107Epoch 00088: val_loss did not improve
1712/1712 [==============================] - 2s 975us/step - loss: 8.6766e-04 - acc: 0.8107 - val_loss: 0.0013 - val_acc: 0.7593
Epoch 89/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.5656e-04 - acc: 0.8172Epoch 00089: val_loss did not improve
1712/1712 [==============================] - 2s 972us/step - loss: 8.5689e-04 - acc: 0.8172 - val_loss: 0.0013 - val_acc: 0.7640
Epoch 90/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.6496e-04 - acc: 0.8178Epoch 00090: val_loss did not improve
1712/1712 [==============================] - 2s 973us/step - loss: 8.6454e-04 - acc: 0.8172 - val_loss: 0.0014 - val_acc: 0.7850
Epoch 91/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.6726e-04 - acc: 0.8231Epoch 00091: val_loss did not improve
1712/1712 [==============================] - 2s 976us/step - loss: 8.6593e-04 - acc: 0.8207 - val_loss: 0.0014 - val_acc: 0.7687
Epoch 92/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.6220e-04 - acc: 0.8249Epoch 00092: val_loss did not improve
1712/1712 [==============================] - 2s 965us/step - loss: 8.6278e-04 - acc: 0.8248 - val_loss: 0.0013 - val_acc: 0.7780
Epoch 93/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.6136e-04 - acc: 0.8125Epoch 00093: val_loss did not improve
1712/1712 [==============================] - 2s 975us/step - loss: 8.6093e-04 - acc: 0.8113 - val_loss: 0.0014 - val_acc: 0.7593
Epoch 94/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.7962e-04 - acc: 0.8096Epoch 00094: val_loss did not improve
1712/1712 [==============================] - 2s 975us/step - loss: 8.7868e-04 - acc: 0.8096 - val_loss: 0.0013 - val_acc: 0.7687
Epoch 95/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.7595e-04 - acc: 0.8149Epoch 00095: val_loss did not improve
1712/1712 [==============================] - 2s 974us/step - loss: 8.7522e-04 - acc: 0.8160 - val_loss: 0.0014 - val_acc: 0.7640
Epoch 96/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.5597e-04 - acc: 0.8013Epoch 00096: val_loss did not improve
1712/1712 [==============================] - 2s 975us/step - loss: 8.5440e-04 - acc: 0.8014 - val_loss: 0.0014 - val_acc: 0.7687
Epoch 97/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.7796e-04 - acc: 0.8072Epoch 00097: val_loss did not improve
1712/1712 [==============================] - 2s 973us/step - loss: 8.7568e-04 - acc: 0.8067 - val_loss: 0.0014 - val_acc: 0.7617
Epoch 98/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.4289e-04 - acc: 0.8143Epoch 00098: val_loss did not improve
1712/1712 [==============================] - 2s 971us/step - loss: 8.4104e-04 - acc: 0.8154 - val_loss: 0.0013 - val_acc: 0.7827
Epoch 99/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.8348e-04 - acc: 0.8031Epoch 00099: val_loss did not improve
1712/1712 [==============================] - 2s 977us/step - loss: 8.8243e-04 - acc: 0.8032 - val_loss: 0.0014 - val_acc: 0.7804
Epoch 100/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.8108e-04 - acc: 0.8154Epoch 00100: val_loss did not improve
1712/1712 [==============================] - 2s 979us/step - loss: 8.8062e-04 - acc: 0.8160 - val_loss: 0.0013 - val_acc: 0.7757


 2017-12-11 23:37:44.384281
Using optimizer:  SGD
Did not load old weights...
Train on 1712 samples, validate on 428 samples
Epoch 1/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.4461e-04 - acc: 0.8184Epoch 00001: val_loss improved from inf to 0.00133, saving model to weights/modelweights_SGD.h5
1712/1712 [==============================] - 2s 1ms/step - loss: 8.4273e-04 - acc: 0.8178 - val_loss: 0.0013 - val_acc: 0.7734
Epoch 2/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.3833e-04 - acc: 0.8178Epoch 00002: val_loss improved from 0.00133 to 0.00133, saving model to weights/modelweights_SGD.h5
1712/1712 [==============================] - 2s 970us/step - loss: 8.3753e-04 - acc: 0.8172 - val_loss: 0.0013 - val_acc: 0.7734
Epoch 3/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.2305e-04 - acc: 0.8166Epoch 00003: val_loss improved from 0.00133 to 0.00133, saving model to weights/modelweights_SGD.h5
1712/1712 [==============================] - 2s 972us/step - loss: 8.2380e-04 - acc: 0.8160 - val_loss: 0.0013 - val_acc: 0.7734
Epoch 4/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.1580e-04 - acc: 0.8137Epoch 00004: val_loss improved from 0.00133 to 0.00132, saving model to weights/modelweights_SGD.h5
1712/1712 [==============================] - 2s 969us/step - loss: 8.1630e-04 - acc: 0.8137 - val_loss: 0.0013 - val_acc: 0.7710
Epoch 5/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.3076e-04 - acc: 0.8178Epoch 00005: val_loss improved from 0.00132 to 0.00132, saving model to weights/modelweights_SGD.h5
1712/1712 [==============================] - 2s 963us/step - loss: 8.3135e-04 - acc: 0.8178 - val_loss: 0.0013 - val_acc: 0.7687
Epoch 6/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.0380e-04 - acc: 0.8048Epoch 00006: val_loss improved from 0.00132 to 0.00132, saving model to weights/modelweights_SGD.h5
1712/1712 [==============================] - 2s 963us/step - loss: 8.0477e-04 - acc: 0.8049 - val_loss: 0.0013 - val_acc: 0.7687
Epoch 7/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.1938e-04 - acc: 0.8196Epoch 00007: val_loss improved from 0.00132 to 0.00132, saving model to weights/modelweights_SGD.h5
1712/1712 [==============================] - 2s 971us/step - loss: 8.1966e-04 - acc: 0.8172 - val_loss: 0.0013 - val_acc: 0.7687
Epoch 8/100
1696/1712 [============================>.] - ETA: 0s - loss: 7.8805e-04 - acc: 0.8208Epoch 00008: val_loss improved from 0.00132 to 0.00132, saving model to weights/modelweights_SGD.h5
1712/1712 [==============================] - 2s 969us/step - loss: 7.8798e-04 - acc: 0.8201 - val_loss: 0.0013 - val_acc: 0.7687
Epoch 9/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.1981e-04 - acc: 0.8166Epoch 00009: val_loss improved from 0.00132 to 0.00132, saving model to weights/modelweights_SGD.h5
1712/1712 [==============================] - 2s 964us/step - loss: 8.1992e-04 - acc: 0.8154 - val_loss: 0.0013 - val_acc: 0.7687
Epoch 10/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.2578e-04 - acc: 0.8208Epoch 00010: val_loss improved from 0.00132 to 0.00131, saving model to weights/modelweights_SGD.h5
1712/1712 [==============================] - 2s 969us/step - loss: 8.2481e-04 - acc: 0.8213 - val_loss: 0.0013 - val_acc: 0.7687
Epoch 11/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.1742e-04 - acc: 0.8278Epoch 00011: val_loss improved from 0.00131 to 0.00131, saving model to weights/modelweights_SGD.h5
1712/1712 [==============================] - 2s 968us/step - loss: 8.1592e-04 - acc: 0.8289 - val_loss: 0.0013 - val_acc: 0.7710
Epoch 12/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.2931e-04 - acc: 0.8107Epoch 00012: val_loss improved from 0.00131 to 0.00131, saving model to weights/modelweights_SGD.h5
1712/1712 [==============================] - 2s 967us/step - loss: 8.2764e-04 - acc: 0.8107 - val_loss: 0.0013 - val_acc: 0.7710
Epoch 13/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.1298e-04 - acc: 0.8172Epoch 00013: val_loss improved from 0.00131 to 0.00131, saving model to weights/modelweights_SGD.h5
1712/1712 [==============================] - 2s 965us/step - loss: 8.1405e-04 - acc: 0.8166 - val_loss: 0.0013 - val_acc: 0.7687
Epoch 14/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.1118e-04 - acc: 0.8101Epoch 00014: val_loss improved from 0.00131 to 0.00131, saving model to weights/modelweights_SGD.h5
1712/1712 [==============================] - 2s 966us/step - loss: 8.0871e-04 - acc: 0.8113 - val_loss: 0.0013 - val_acc: 0.7687
Epoch 15/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.2113e-04 - acc: 0.8072Epoch 00015: val_loss improved from 0.00131 to 0.00131, saving model to weights/modelweights_SGD.h5
1712/1712 [==============================] - 2s 963us/step - loss: 8.2041e-04 - acc: 0.8078 - val_loss: 0.0013 - val_acc: 0.7710
Epoch 16/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.3608e-04 - acc: 0.8084Epoch 00016: val_loss improved from 0.00131 to 0.00131, saving model to weights/modelweights_SGD.h5
1712/1712 [==============================] - 2s 966us/step - loss: 8.3570e-04 - acc: 0.8067 - val_loss: 0.0013 - val_acc: 0.7710
Epoch 17/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.0812e-04 - acc: 0.8066Epoch 00017: val_loss improved from 0.00131 to 0.00131, saving model to weights/modelweights_SGD.h5
1712/1712 [==============================] - 2s 967us/step - loss: 8.0757e-04 - acc: 0.8072 - val_loss: 0.0013 - val_acc: 0.7710
Epoch 18/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.2046e-04 - acc: 0.8125Epoch 00018: val_loss improved from 0.00131 to 0.00131, saving model to weights/modelweights_SGD.h5
1712/1712 [==============================] - 2s 966us/step - loss: 8.2046e-04 - acc: 0.8131 - val_loss: 0.0013 - val_acc: 0.7664
Epoch 19/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.0649e-04 - acc: 0.8096Epoch 00019: val_loss improved from 0.00131 to 0.00131, saving model to weights/modelweights_SGD.h5
1712/1712 [==============================] - 2s 961us/step - loss: 8.0476e-04 - acc: 0.8078 - val_loss: 0.0013 - val_acc: 0.7687
Epoch 20/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.0087e-04 - acc: 0.8166Epoch 00020: val_loss improved from 0.00131 to 0.00131, saving model to weights/modelweights_SGD.h5
1712/1712 [==============================] - 2s 964us/step - loss: 8.0020e-04 - acc: 0.8166 - val_loss: 0.0013 - val_acc: 0.7687
Epoch 21/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.1697e-04 - acc: 0.8107Epoch 00021: val_loss did not improve
1712/1712 [==============================] - 2s 927us/step - loss: 8.1671e-04 - acc: 0.8125 - val_loss: 0.0013 - val_acc: 0.7687
Epoch 22/100
1696/1712 [============================>.] - ETA: 0s - loss: 7.9746e-04 - acc: 0.8143Epoch 00022: val_loss improved from 0.00131 to 0.00131, saving model to weights/modelweights_SGD.h5
1712/1712 [==============================] - 2s 972us/step - loss: 7.9815e-04 - acc: 0.8143 - val_loss: 0.0013 - val_acc: 0.7687
Epoch 23/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.1880e-04 - acc: 0.8308Epoch 00023: val_loss improved from 0.00131 to 0.00131, saving model to weights/modelweights_SGD.h5
1712/1712 [==============================] - 2s 967us/step - loss: 8.1782e-04 - acc: 0.8294 - val_loss: 0.0013 - val_acc: 0.7710
Epoch 24/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.2906e-04 - acc: 0.8019Epoch 00024: val_loss did not improve
1712/1712 [==============================] - 2s 924us/step - loss: 8.2840e-04 - acc: 0.8032 - val_loss: 0.0013 - val_acc: 0.7687
Epoch 25/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.0537e-04 - acc: 0.8054Epoch 00025: val_loss improved from 0.00131 to 0.00131, saving model to weights/modelweights_SGD.h5
1712/1712 [==============================] - 2s 963us/step - loss: 8.0472e-04 - acc: 0.8067 - val_loss: 0.0013 - val_acc: 0.7734
Epoch 26/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.1025e-04 - acc: 0.8054Epoch 00026: val_loss improved from 0.00131 to 0.00131, saving model to weights/modelweights_SGD.h5
1712/1712 [==============================] - 2s 972us/step - loss: 8.0899e-04 - acc: 0.8061 - val_loss: 0.0013 - val_acc: 0.7710
Epoch 27/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.1379e-04 - acc: 0.8101Epoch 00027: val_loss improved from 0.00131 to 0.00131, saving model to weights/modelweights_SGD.h5
1712/1712 [==============================] - 2s 966us/step - loss: 8.1752e-04 - acc: 0.8096 - val_loss: 0.0013 - val_acc: 0.7710
Epoch 28/100
1696/1712 [============================>.] - ETA: 0s - loss: 7.9327e-04 - acc: 0.8131Epoch 00028: val_loss improved from 0.00131 to 0.00131, saving model to weights/modelweights_SGD.h5
1712/1712 [==============================] - 2s 967us/step - loss: 7.9730e-04 - acc: 0.8113 - val_loss: 0.0013 - val_acc: 0.7710
Epoch 29/100
1696/1712 [============================>.] - ETA: 0s - loss: 7.9975e-04 - acc: 0.8261Epoch 00029: val_loss improved from 0.00131 to 0.00131, saving model to weights/modelweights_SGD.h5
1712/1712 [==============================] - 2s 966us/step - loss: 7.9960e-04 - acc: 0.8254 - val_loss: 0.0013 - val_acc: 0.7687
Epoch 30/100
1696/1712 [============================>.] - ETA: 0s - loss: 7.9475e-04 - acc: 0.8066Epoch 00030: val_loss improved from 0.00131 to 0.00131, saving model to weights/modelweights_SGD.h5
1712/1712 [==============================] - 2s 977us/step - loss: 7.9306e-04 - acc: 0.8072 - val_loss: 0.0013 - val_acc: 0.7710
Epoch 31/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.2636e-04 - acc: 0.8172Epoch 00031: val_loss improved from 0.00131 to 0.00131, saving model to weights/modelweights_SGD.h5
1712/1712 [==============================] - 2s 972us/step - loss: 8.2605e-04 - acc: 0.8160 - val_loss: 0.0013 - val_acc: 0.7734
Epoch 32/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.0284e-04 - acc: 0.8013Epoch 00032: val_loss improved from 0.00131 to 0.00131, saving model to weights/modelweights_SGD.h5
1712/1712 [==============================] - 2s 971us/step - loss: 8.0133e-04 - acc: 0.8020 - val_loss: 0.0013 - val_acc: 0.7734
Epoch 33/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.0967e-04 - acc: 0.8154Epoch 00033: val_loss improved from 0.00131 to 0.00131, saving model to weights/modelweights_SGD.h5
1712/1712 [==============================] - 2s 968us/step - loss: 8.0987e-04 - acc: 0.8166 - val_loss: 0.0013 - val_acc: 0.7710
Epoch 34/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.0939e-04 - acc: 0.8278Epoch 00034: val_loss improved from 0.00131 to 0.00131, saving model to weights/modelweights_SGD.h5
1712/1712 [==============================] - 2s 968us/step - loss: 8.0728e-04 - acc: 0.8271 - val_loss: 0.0013 - val_acc: 0.7710
Epoch 35/100
1696/1712 [============================>.] - ETA: 0s - loss: 7.9724e-04 - acc: 0.8149Epoch 00035: val_loss improved from 0.00131 to 0.00131, saving model to weights/modelweights_SGD.h5
1712/1712 [==============================] - 2s 970us/step - loss: 7.9643e-04 - acc: 0.8137 - val_loss: 0.0013 - val_acc: 0.7710
Epoch 36/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.2148e-04 - acc: 0.8101Epoch 00036: val_loss improved from 0.00131 to 0.00131, saving model to weights/modelweights_SGD.h5
1712/1712 [==============================] - 2s 971us/step - loss: 8.2279e-04 - acc: 0.8096 - val_loss: 0.0013 - val_acc: 0.7710
Epoch 37/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.0424e-04 - acc: 0.8149Epoch 00037: val_loss improved from 0.00131 to 0.00131, saving model to weights/modelweights_SGD.h5
1712/1712 [==============================] - 2s 968us/step - loss: 8.0370e-04 - acc: 0.8166 - val_loss: 0.0013 - val_acc: 0.7710
Epoch 38/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.2082e-04 - acc: 0.8066Epoch 00038: val_loss improved from 0.00131 to 0.00130, saving model to weights/modelweights_SGD.h5
1712/1712 [==============================] - 2s 971us/step - loss: 8.2148e-04 - acc: 0.8061 - val_loss: 0.0013 - val_acc: 0.7710
Epoch 39/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.0511e-04 - acc: 0.7978Epoch 00039: val_loss did not improve
1712/1712 [==============================] - 2s 930us/step - loss: 8.0355e-04 - acc: 0.7991 - val_loss: 0.0013 - val_acc: 0.7710
Epoch 40/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.0282e-04 - acc: 0.8060Epoch 00040: val_loss improved from 0.00130 to 0.00130, saving model to weights/modelweights_SGD.h5
1712/1712 [==============================] - 2s 967us/step - loss: 8.0198e-04 - acc: 0.8061 - val_loss: 0.0013 - val_acc: 0.7734
Epoch 41/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.0620e-04 - acc: 0.8096Epoch 00041: val_loss did not improve
1712/1712 [==============================] - 2s 929us/step - loss: 8.0780e-04 - acc: 0.8090 - val_loss: 0.0013 - val_acc: 0.7710
Epoch 42/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.0014e-04 - acc: 0.8131Epoch 00042: val_loss improved from 0.00130 to 0.00130, saving model to weights/modelweights_SGD.h5
1712/1712 [==============================] - 2s 969us/step - loss: 7.9849e-04 - acc: 0.8137 - val_loss: 0.0013 - val_acc: 0.7734
Epoch 43/100
1696/1712 [============================>.] - ETA: 0s - loss: 7.9663e-04 - acc: 0.8149Epoch 00043: val_loss did not improve
1712/1712 [==============================] - 2s 928us/step - loss: 7.9493e-04 - acc: 0.8131 - val_loss: 0.0013 - val_acc: 0.7710
Epoch 44/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.0187e-04 - acc: 0.8160Epoch 00044: val_loss did not improve
1712/1712 [==============================] - 2s 933us/step - loss: 8.0277e-04 - acc: 0.8154 - val_loss: 0.0013 - val_acc: 0.7687
Epoch 45/100
1696/1712 [============================>.] - ETA: 0s - loss: 7.9971e-04 - acc: 0.8143Epoch 00045: val_loss improved from 0.00130 to 0.00130, saving model to weights/modelweights_SGD.h5
1712/1712 [==============================] - 2s 970us/step - loss: 7.9783e-04 - acc: 0.8131 - val_loss: 0.0013 - val_acc: 0.7710
Epoch 46/100
1696/1712 [============================>.] - ETA: 0s - loss: 7.8893e-04 - acc: 0.8125Epoch 00046: val_loss did not improve
1712/1712 [==============================] - 2s 934us/step - loss: 7.9014e-04 - acc: 0.8131 - val_loss: 0.0013 - val_acc: 0.7687
Epoch 47/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.0366e-04 - acc: 0.8054Epoch 00047: val_loss did not improve
1712/1712 [==============================] - 2s 932us/step - loss: 8.0334e-04 - acc: 0.8037 - val_loss: 0.0013 - val_acc: 0.7687
Epoch 48/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.2318e-04 - acc: 0.8096Epoch 00048: val_loss improved from 0.00130 to 0.00130, saving model to weights/modelweights_SGD.h5
1712/1712 [==============================] - 2s 964us/step - loss: 8.2142e-04 - acc: 0.8102 - val_loss: 0.0013 - val_acc: 0.7710
Epoch 49/100
1696/1712 [============================>.] - ETA: 0s - loss: 7.9487e-04 - acc: 0.8031Epoch 00049: val_loss improved from 0.00130 to 0.00130, saving model to weights/modelweights_SGD.h5
1712/1712 [==============================] - 2s 968us/step - loss: 7.9325e-04 - acc: 0.8026 - val_loss: 0.0013 - val_acc: 0.7710
Epoch 50/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.0938e-04 - acc: 0.8107Epoch 00050: val_loss did not improve
1712/1712 [==============================] - 2s 933us/step - loss: 8.0803e-04 - acc: 0.8125 - val_loss: 0.0013 - val_acc: 0.7687
Epoch 51/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.0561e-04 - acc: 0.8137Epoch 00051: val_loss improved from 0.00130 to 0.00130, saving model to weights/modelweights_SGD.h5
1712/1712 [==============================] - 2s 966us/step - loss: 8.0469e-04 - acc: 0.8137 - val_loss: 0.0013 - val_acc: 0.7687
Epoch 52/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.0222e-04 - acc: 0.8296Epoch 00052: val_loss improved from 0.00130 to 0.00130, saving model to weights/modelweights_SGD.h5
1712/1712 [==============================] - 2s 970us/step - loss: 8.0137e-04 - acc: 0.8300 - val_loss: 0.0013 - val_acc: 0.7687
Epoch 53/100
1696/1712 [============================>.] - ETA: 0s - loss: 7.9795e-04 - acc: 0.8160Epoch 00053: val_loss improved from 0.00130 to 0.00130, saving model to weights/modelweights_SGD.h5
1712/1712 [==============================] - 2s 974us/step - loss: 7.9767e-04 - acc: 0.8166 - val_loss: 0.0013 - val_acc: 0.7687
Epoch 54/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.0809e-04 - acc: 0.7995Epoch 00054: val_loss did not improve
1712/1712 [==============================] - 2s 938us/step - loss: 8.1002e-04 - acc: 0.7996 - val_loss: 0.0013 - val_acc: 0.7687
Epoch 55/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.0127e-04 - acc: 0.8154Epoch 00055: val_loss did not improve
1712/1712 [==============================] - 2s 937us/step - loss: 8.0129e-04 - acc: 0.8154 - val_loss: 0.0013 - val_acc: 0.7687
Epoch 56/100
1696/1712 [============================>.] - ETA: 0s - loss: 7.8898e-04 - acc: 0.8025Epoch 00056: val_loss improved from 0.00130 to 0.00130, saving model to weights/modelweights_SGD.h5
1712/1712 [==============================] - 2s 965us/step - loss: 7.8842e-04 - acc: 0.8037 - val_loss: 0.0013 - val_acc: 0.7664
Epoch 57/100
1696/1712 [============================>.] - ETA: 0s - loss: 7.9899e-04 - acc: 0.8125Epoch 00057: val_loss did not improve
1712/1712 [==============================] - 2s 933us/step - loss: 8.0431e-04 - acc: 0.8131 - val_loss: 0.0013 - val_acc: 0.7640
Epoch 58/100
1696/1712 [============================>.] - ETA: 0s - loss: 7.9822e-04 - acc: 0.8066Epoch 00058: val_loss did not improve
1712/1712 [==============================] - 2s 930us/step - loss: 7.9932e-04 - acc: 0.8067 - val_loss: 0.0013 - val_acc: 0.7687
Epoch 59/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.1219e-04 - acc: 0.8143Epoch 00059: val_loss improved from 0.00130 to 0.00130, saving model to weights/modelweights_SGD.h5
1712/1712 [==============================] - 2s 1ms/step - loss: 8.1008e-04 - acc: 0.8143 - val_loss: 0.0013 - val_acc: 0.7664
Epoch 60/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.1302e-04 - acc: 0.8202Epoch 00060: val_loss improved from 0.00130 to 0.00130, saving model to weights/modelweights_SGD.h5
1712/1712 [==============================] - 2s 969us/step - loss: 8.1387e-04 - acc: 0.8201 - val_loss: 0.0013 - val_acc: 0.7687
Epoch 61/100
1696/1712 [============================>.] - ETA: 0s - loss: 7.9853e-04 - acc: 0.8231Epoch 00061: val_loss improved from 0.00130 to 0.00130, saving model to weights/modelweights_SGD.h5
1712/1712 [==============================] - 2s 967us/step - loss: 7.9715e-04 - acc: 0.8236 - val_loss: 0.0013 - val_acc: 0.7664
Epoch 62/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.1051e-04 - acc: 0.8178Epoch 00062: val_loss did not improve
1712/1712 [==============================] - 2s 933us/step - loss: 8.0986e-04 - acc: 0.8166 - val_loss: 0.0013 - val_acc: 0.7664
Epoch 63/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.1183e-04 - acc: 0.8078Epoch 00063: val_loss improved from 0.00130 to 0.00130, saving model to weights/modelweights_SGD.h5
1712/1712 [==============================] - 2s 967us/step - loss: 8.1209e-04 - acc: 0.8061 - val_loss: 0.0013 - val_acc: 0.7664
Epoch 64/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.0196e-04 - acc: 0.8184Epoch 00064: val_loss did not improve
1712/1712 [==============================] - 2s 934us/step - loss: 8.0279e-04 - acc: 0.8189 - val_loss: 0.0013 - val_acc: 0.7664
Epoch 65/100
1696/1712 [============================>.] - ETA: 0s - loss: 7.8253e-04 - acc: 0.7983Epoch 00065: val_loss did not improve
1712/1712 [==============================] - 2s 931us/step - loss: 7.8629e-04 - acc: 0.7985 - val_loss: 0.0013 - val_acc: 0.7664
Epoch 66/100
1696/1712 [============================>.] - ETA: 0s - loss: 7.7667e-04 - acc: 0.8042Epoch 00066: val_loss did not improve
1712/1712 [==============================] - 2s 932us/step - loss: 7.7729e-04 - acc: 0.8049 - val_loss: 0.0013 - val_acc: 0.7664
Epoch 67/100
1696/1712 [============================>.] - ETA: 0s - loss: 7.9620e-04 - acc: 0.8125Epoch 00067: val_loss did not improve
1712/1712 [==============================] - 2s 936us/step - loss: 7.9604e-04 - acc: 0.8125 - val_loss: 0.0013 - val_acc: 0.7664
Epoch 68/100
1696/1712 [============================>.] - ETA: 0s - loss: 7.9254e-04 - acc: 0.8154Epoch 00068: val_loss improved from 0.00130 to 0.00130, saving model to weights/modelweights_SGD.h5
1712/1712 [==============================] - 2s 969us/step - loss: 7.9341e-04 - acc: 0.8148 - val_loss: 0.0013 - val_acc: 0.7664
Epoch 69/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.0464e-04 - acc: 0.8125Epoch 00069: val_loss improved from 0.00130 to 0.00130, saving model to weights/modelweights_SGD.h5
1712/1712 [==============================] - 2s 967us/step - loss: 8.0268e-04 - acc: 0.8131 - val_loss: 0.0013 - val_acc: 0.7664
Epoch 70/100
1696/1712 [============================>.] - ETA: 0s - loss: 7.8753e-04 - acc: 0.8184Epoch 00070: val_loss did not improve
1712/1712 [==============================] - 2s 932us/step - loss: 7.8691e-04 - acc: 0.8189 - val_loss: 0.0013 - val_acc: 0.7664
Epoch 71/100
1696/1712 [============================>.] - ETA: 0s - loss: 7.9986e-04 - acc: 0.8037Epoch 00071: val_loss improved from 0.00130 to 0.00130, saving model to weights/modelweights_SGD.h5
1712/1712 [==============================] - 2s 972us/step - loss: 7.9816e-04 - acc: 0.8037 - val_loss: 0.0013 - val_acc: 0.7664
Epoch 72/100
1696/1712 [============================>.] - ETA: 0s - loss: 7.9185e-04 - acc: 0.8143Epoch 00072: val_loss did not improve
1712/1712 [==============================] - 2s 936us/step - loss: 7.9526e-04 - acc: 0.8160 - val_loss: 0.0013 - val_acc: 0.7640
Epoch 73/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.1316e-04 - acc: 0.8119Epoch 00073: val_loss did not improve
1712/1712 [==============================] - 2s 932us/step - loss: 8.1109e-04 - acc: 0.8107 - val_loss: 0.0013 - val_acc: 0.7640
Epoch 74/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.0070e-04 - acc: 0.8031Epoch 00074: val_loss did not improve
1712/1712 [==============================] - 2s 937us/step - loss: 7.9882e-04 - acc: 0.8026 - val_loss: 0.0013 - val_acc: 0.7640
Epoch 75/100
1696/1712 [============================>.] - ETA: 0s - loss: 7.9074e-04 - acc: 0.8131Epoch 00075: val_loss did not improve
1712/1712 [==============================] - 2s 935us/step - loss: 7.9015e-04 - acc: 0.8137 - val_loss: 0.0013 - val_acc: 0.7640
Epoch 76/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.0133e-04 - acc: 0.8101Epoch 00076: val_loss did not improve
1712/1712 [==============================] - 2s 928us/step - loss: 7.9903e-04 - acc: 0.8113 - val_loss: 0.0013 - val_acc: 0.7640
Epoch 77/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.1666e-04 - acc: 0.8160Epoch 00077: val_loss did not improve
1712/1712 [==============================] - 2s 925us/step - loss: 8.1518e-04 - acc: 0.8148 - val_loss: 0.0013 - val_acc: 0.7664
Epoch 78/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.1010e-04 - acc: 0.8048Epoch 00078: val_loss did not improve
1712/1712 [==============================] - 2s 930us/step - loss: 8.1224e-04 - acc: 0.8067 - val_loss: 0.0013 - val_acc: 0.7664
Epoch 79/100
1696/1712 [============================>.] - ETA: 0s - loss: 7.9121e-04 - acc: 0.8019Epoch 00079: val_loss improved from 0.00130 to 0.00130, saving model to weights/modelweights_SGD.h5
1712/1712 [==============================] - 2s 964us/step - loss: 7.9117e-04 - acc: 0.8037 - val_loss: 0.0013 - val_acc: 0.7664
Epoch 80/100
1696/1712 [============================>.] - ETA: 0s - loss: 7.9498e-04 - acc: 0.8113Epoch 00080: val_loss improved from 0.00130 to 0.00130, saving model to weights/modelweights_SGD.h5
1712/1712 [==============================] - 2s 969us/step - loss: 7.9527e-04 - acc: 0.8107 - val_loss: 0.0013 - val_acc: 0.7664
Epoch 81/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.0379e-04 - acc: 0.8037Epoch 00081: val_loss improved from 0.00130 to 0.00130, saving model to weights/modelweights_SGD.h5
1712/1712 [==============================] - 2s 970us/step - loss: 8.0292e-04 - acc: 0.8032 - val_loss: 0.0013 - val_acc: 0.7664
Epoch 82/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.0854e-04 - acc: 0.8137Epoch 00082: val_loss did not improve
1712/1712 [==============================] - 2s 936us/step - loss: 8.0919e-04 - acc: 0.8131 - val_loss: 0.0013 - val_acc: 0.7664
Epoch 83/100
1696/1712 [============================>.] - ETA: 0s - loss: 7.9908e-04 - acc: 0.8196Epoch 00083: val_loss improved from 0.00130 to 0.00130, saving model to weights/modelweights_SGD.h5
1712/1712 [==============================] - 2s 972us/step - loss: 7.9836e-04 - acc: 0.8189 - val_loss: 0.0013 - val_acc: 0.7664
Epoch 84/100
1696/1712 [============================>.] - ETA: 0s - loss: 7.8053e-04 - acc: 0.8037Epoch 00084: val_loss did not improve
1712/1712 [==============================] - 2s 933us/step - loss: 7.8053e-04 - acc: 0.8043 - val_loss: 0.0013 - val_acc: 0.7664
Epoch 85/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.1793e-04 - acc: 0.8113Epoch 00085: val_loss did not improve
1712/1712 [==============================] - 2s 931us/step - loss: 8.1465e-04 - acc: 0.8113 - val_loss: 0.0013 - val_acc: 0.7640
Epoch 86/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.0963e-04 - acc: 0.8042Epoch 00086: val_loss did not improve
1712/1712 [==============================] - 2s 926us/step - loss: 8.1269e-04 - acc: 0.8032 - val_loss: 0.0013 - val_acc: 0.7664
Epoch 87/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.0515e-04 - acc: 0.8096Epoch 00087: val_loss did not improve
1712/1712 [==============================] - 2s 933us/step - loss: 8.0801e-04 - acc: 0.8107 - val_loss: 0.0013 - val_acc: 0.7664
Epoch 88/100
1696/1712 [============================>.] - ETA: 0s - loss: 7.9006e-04 - acc: 0.8042Epoch 00088: val_loss did not improve
1712/1712 [==============================] - 2s 928us/step - loss: 7.9037e-04 - acc: 0.8049 - val_loss: 0.0013 - val_acc: 0.7664
Epoch 89/100
1696/1712 [============================>.] - ETA: 0s - loss: 7.8879e-04 - acc: 0.8154Epoch 00089: val_loss did not improve
1712/1712 [==============================] - 2s 932us/step - loss: 7.8902e-04 - acc: 0.8148 - val_loss: 0.0013 - val_acc: 0.7664
Epoch 90/100
1696/1712 [============================>.] - ETA: 0s - loss: 7.9016e-04 - acc: 0.8190Epoch 00090: val_loss improved from 0.00130 to 0.00130, saving model to weights/modelweights_SGD.h5
1712/1712 [==============================] - 2s 968us/step - loss: 7.9191e-04 - acc: 0.8195 - val_loss: 0.0013 - val_acc: 0.7664
Epoch 91/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.1594e-04 - acc: 0.8037Epoch 00091: val_loss improved from 0.00130 to 0.00130, saving model to weights/modelweights_SGD.h5
1712/1712 [==============================] - 2s 971us/step - loss: 8.1544e-04 - acc: 0.8049 - val_loss: 0.0013 - val_acc: 0.7664
Epoch 92/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.0001e-04 - acc: 0.8149Epoch 00092: val_loss improved from 0.00130 to 0.00130, saving model to weights/modelweights_SGD.h5
1712/1712 [==============================] - 2s 984us/step - loss: 7.9996e-04 - acc: 0.8154 - val_loss: 0.0013 - val_acc: 0.7664
Epoch 93/100
1696/1712 [============================>.] - ETA: 0s - loss: 7.9990e-04 - acc: 0.8037Epoch 00093: val_loss improved from 0.00130 to 0.00130, saving model to weights/modelweights_SGD.h5
1712/1712 [==============================] - 2s 976us/step - loss: 8.0066e-04 - acc: 0.8032 - val_loss: 0.0013 - val_acc: 0.7664
Epoch 94/100
1696/1712 [============================>.] - ETA: 0s - loss: 7.8548e-04 - acc: 0.8084Epoch 00094: val_loss did not improve
1712/1712 [==============================] - 2s 935us/step - loss: 7.8510e-04 - acc: 0.8084 - val_loss: 0.0013 - val_acc: 0.7664
Epoch 95/100
1696/1712 [============================>.] - ETA: 0s - loss: 7.9556e-04 - acc: 0.8219Epoch 00095: val_loss did not improve
1712/1712 [==============================] - 2s 925us/step - loss: 7.9735e-04 - acc: 0.8207 - val_loss: 0.0013 - val_acc: 0.7664
Epoch 96/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.0900e-04 - acc: 0.8143Epoch 00096: val_loss did not improve
1712/1712 [==============================] - 2s 929us/step - loss: 8.0660e-04 - acc: 0.8143 - val_loss: 0.0013 - val_acc: 0.7664
Epoch 97/100
1696/1712 [============================>.] - ETA: 0s - loss: 7.8369e-04 - acc: 0.8208Epoch 00097: val_loss did not improve
1712/1712 [==============================] - 2s 929us/step - loss: 7.9013e-04 - acc: 0.8195 - val_loss: 0.0013 - val_acc: 0.7664
Epoch 98/100
1696/1712 [============================>.] - ETA: 0s - loss: 7.9630e-04 - acc: 0.8078Epoch 00098: val_loss did not improve
1712/1712 [==============================] - 2s 928us/step - loss: 7.9550e-04 - acc: 0.8078 - val_loss: 0.0013 - val_acc: 0.7687
Epoch 99/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.1296e-04 - acc: 0.8107Epoch 00099: val_loss did not improve
1712/1712 [==============================] - 2s 933us/step - loss: 8.1193e-04 - acc: 0.8107 - val_loss: 0.0013 - val_acc: 0.7664
Epoch 100/100
1696/1712 [============================>.] - ETA: 0s - loss: 7.9286e-04 - acc: 0.8178Epoch 00100: val_loss did not improve
1712/1712 [==============================] - 2s 935us/step - loss: 7.9086e-04 - acc: 0.8160 - val_loss: 0.0013 - val_acc: 0.7664


 2017-12-11 23:40:29.014980
Using optimizer:  Adam
Did not load old weights...
Train on 1712 samples, validate on 428 samples
Epoch 1/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.7042e-04 - acc: 0.8096Epoch 00001: val_loss improved from inf to 0.00133, saving model to weights/modelweights_Adam.h5
1712/1712 [==============================] - 3s 2ms/step - loss: 8.7266e-04 - acc: 0.8090 - val_loss: 0.0013 - val_acc: 0.7640
Epoch 2/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.3160e-04 - acc: 0.8131Epoch 00002: val_loss improved from 0.00133 to 0.00132, saving model to weights/modelweights_Adam.h5
1712/1712 [==============================] - 2s 1ms/step - loss: 8.3199e-04 - acc: 0.8131 - val_loss: 0.0013 - val_acc: 0.7687
Epoch 3/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.5565e-04 - acc: 0.8101Epoch 00003: val_loss did not improve
1712/1712 [==============================] - 2s 1ms/step - loss: 8.5399e-04 - acc: 0.8119 - val_loss: 0.0013 - val_acc: 0.7780
Epoch 4/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.5827e-04 - acc: 0.8190Epoch 00004: val_loss did not improve
1712/1712 [==============================] - 2s 1ms/step - loss: 8.5777e-04 - acc: 0.8189 - val_loss: 0.0013 - val_acc: 0.7804
Epoch 5/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.1631e-04 - acc: 0.8060Epoch 00005: val_loss did not improve
1712/1712 [==============================] - 2s 1ms/step - loss: 8.1875e-04 - acc: 0.8049 - val_loss: 0.0013 - val_acc: 0.7687
Epoch 6/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.2787e-04 - acc: 0.8031Epoch 00006: val_loss improved from 0.00132 to 0.00132, saving model to weights/modelweights_Adam.h5
1712/1712 [==============================] - 2s 1ms/step - loss: 8.2971e-04 - acc: 0.8020 - val_loss: 0.0013 - val_acc: 0.7640
Epoch 7/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.3365e-04 - acc: 0.8190Epoch 00007: val_loss did not improve
1712/1712 [==============================] - 2s 1ms/step - loss: 8.3278e-04 - acc: 0.8189 - val_loss: 0.0013 - val_acc: 0.7757
Epoch 8/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.4647e-04 - acc: 0.8154Epoch 00008: val_loss improved from 0.00132 to 0.00131, saving model to weights/modelweights_Adam.h5
1712/1712 [==============================] - 2s 1ms/step - loss: 8.4503e-04 - acc: 0.8154 - val_loss: 0.0013 - val_acc: 0.7710
Epoch 9/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.5056e-04 - acc: 0.8119Epoch 00009: val_loss did not improve
1712/1712 [==============================] - 2s 1ms/step - loss: 8.5079e-04 - acc: 0.8113 - val_loss: 0.0014 - val_acc: 0.7734
Epoch 10/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.5568e-04 - acc: 0.8113Epoch 00010: val_loss did not improve
1712/1712 [==============================] - 2s 1ms/step - loss: 8.5679e-04 - acc: 0.8102 - val_loss: 0.0013 - val_acc: 0.7710
Epoch 11/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.5682e-04 - acc: 0.8084Epoch 00011: val_loss did not improve
1712/1712 [==============================] - 2s 1ms/step - loss: 8.5513e-04 - acc: 0.8096 - val_loss: 0.0013 - val_acc: 0.7710
Epoch 12/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.4257e-04 - acc: 0.8160Epoch 00012: val_loss did not improve
1712/1712 [==============================] - 2s 1ms/step - loss: 8.4145e-04 - acc: 0.8160 - val_loss: 0.0013 - val_acc: 0.7804
Epoch 13/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.3629e-04 - acc: 0.8125Epoch 00013: val_loss did not improve
1712/1712 [==============================] - 2s 1ms/step - loss: 8.3355e-04 - acc: 0.8137 - val_loss: 0.0013 - val_acc: 0.7523
Epoch 14/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.3871e-04 - acc: 0.8019Epoch 00014: val_loss did not improve
1712/1712 [==============================] - 2s 1ms/step - loss: 8.3930e-04 - acc: 0.8014 - val_loss: 0.0014 - val_acc: 0.7780
Epoch 15/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.2765e-04 - acc: 0.8213Epoch 00015: val_loss did not improve
1712/1712 [==============================] - 2s 1ms/step - loss: 8.2946e-04 - acc: 0.8213 - val_loss: 0.0013 - val_acc: 0.7570
Epoch 16/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.4252e-04 - acc: 0.8054Epoch 00016: val_loss did not improve
1712/1712 [==============================] - 2s 1ms/step - loss: 8.4145e-04 - acc: 0.8055 - val_loss: 0.0014 - val_acc: 0.7827
Epoch 17/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.5053e-04 - acc: 0.8096Epoch 00017: val_loss did not improve
1712/1712 [==============================] - 2s 1ms/step - loss: 8.4795e-04 - acc: 0.8107 - val_loss: 0.0013 - val_acc: 0.7850
Epoch 18/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.6401e-04 - acc: 0.8025Epoch 00018: val_loss did not improve
1712/1712 [==============================] - 2s 1ms/step - loss: 8.6491e-04 - acc: 0.8026 - val_loss: 0.0013 - val_acc: 0.7687
Epoch 19/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.5307e-04 - acc: 0.8255Epoch 00019: val_loss did not improve
1712/1712 [==============================] - 2s 1ms/step - loss: 8.5051e-04 - acc: 0.8248 - val_loss: 0.0013 - val_acc: 0.7547
Epoch 20/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.6379e-04 - acc: 0.8160Epoch 00020: val_loss did not improve
1712/1712 [==============================] - 2s 1ms/step - loss: 8.6319e-04 - acc: 0.8166 - val_loss: 0.0013 - val_acc: 0.7617
Epoch 21/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.4415e-04 - acc: 0.8154Epoch 00021: val_loss did not improve
1712/1712 [==============================] - 2s 1ms/step - loss: 8.4440e-04 - acc: 0.8148 - val_loss: 0.0013 - val_acc: 0.7850
Epoch 22/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.4934e-04 - acc: 0.8125Epoch 00022: val_loss did not improve
1712/1712 [==============================] - 2s 1ms/step - loss: 8.4840e-04 - acc: 0.8125 - val_loss: 0.0014 - val_acc: 0.7640
Epoch 23/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.4243e-04 - acc: 0.8101Epoch 00023: val_loss did not improve
1712/1712 [==============================] - 2s 1ms/step - loss: 8.4310e-04 - acc: 0.8090 - val_loss: 0.0013 - val_acc: 0.7710
Epoch 24/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.4613e-04 - acc: 0.8072Epoch 00024: val_loss did not improve
1712/1712 [==============================] - 2s 1ms/step - loss: 8.4697e-04 - acc: 0.8043 - val_loss: 0.0013 - val_acc: 0.7640
Epoch 25/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.5039e-04 - acc: 0.8178Epoch 00025: val_loss did not improve
1712/1712 [==============================] - 2s 1ms/step - loss: 8.4983e-04 - acc: 0.8189 - val_loss: 0.0013 - val_acc: 0.7664
Epoch 26/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.3628e-04 - acc: 0.8243Epoch 00026: val_loss did not improve
1712/1712 [==============================] - 2s 1ms/step - loss: 8.3472e-04 - acc: 0.8236 - val_loss: 0.0013 - val_acc: 0.7640
Epoch 27/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.3774e-04 - acc: 0.8166Epoch 00027: val_loss did not improve
1712/1712 [==============================] - 2s 1ms/step - loss: 8.3714e-04 - acc: 0.8166 - val_loss: 0.0013 - val_acc: 0.7780
Epoch 28/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.0027e-04 - acc: 0.8107Epoch 00028: val_loss did not improve
1712/1712 [==============================] - 2s 1ms/step - loss: 8.0052e-04 - acc: 0.8107 - val_loss: 0.0014 - val_acc: 0.7734
Epoch 29/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.4488e-04 - acc: 0.8202Epoch 00029: val_loss did not improve
1712/1712 [==============================] - 2s 1ms/step - loss: 8.4354e-04 - acc: 0.8207 - val_loss: 0.0013 - val_acc: 0.7757
Epoch 30/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.5924e-04 - acc: 0.8190Epoch 00030: val_loss did not improve
1712/1712 [==============================] - 2s 1ms/step - loss: 8.5801e-04 - acc: 0.8195 - val_loss: 0.0013 - val_acc: 0.7617
Epoch 31/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.4886e-04 - acc: 0.8060Epoch 00031: val_loss did not improve
1712/1712 [==============================] - 2s 1ms/step - loss: 8.4807e-04 - acc: 0.8055 - val_loss: 0.0013 - val_acc: 0.7617
Epoch 32/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.3921e-04 - acc: 0.8042Epoch 00032: val_loss did not improve
1712/1712 [==============================] - 2s 1ms/step - loss: 8.4122e-04 - acc: 0.8043 - val_loss: 0.0014 - val_acc: 0.7477
Epoch 33/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.4295e-04 - acc: 0.8202Epoch 00033: val_loss did not improve
1712/1712 [==============================] - 2s 1ms/step - loss: 8.4358e-04 - acc: 0.8195 - val_loss: 0.0014 - val_acc: 0.7523
Epoch 34/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.2906e-04 - acc: 0.8125Epoch 00034: val_loss did not improve
1712/1712 [==============================] - 2s 1ms/step - loss: 8.2672e-04 - acc: 0.8119 - val_loss: 0.0014 - val_acc: 0.7687
Epoch 35/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.2575e-04 - acc: 0.8178Epoch 00035: val_loss did not improve
1712/1712 [==============================] - 2s 1ms/step - loss: 8.2696e-04 - acc: 0.8189 - val_loss: 0.0014 - val_acc: 0.7640
Epoch 36/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.3640e-04 - acc: 0.8149Epoch 00036: val_loss did not improve
1712/1712 [==============================] - 2s 1ms/step - loss: 8.3599e-04 - acc: 0.8148 - val_loss: 0.0013 - val_acc: 0.7757
Epoch 37/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.3281e-04 - acc: 0.8213Epoch 00037: val_loss did not improve
1712/1712 [==============================] - 2s 1ms/step - loss: 8.3449e-04 - acc: 0.8218 - val_loss: 0.0013 - val_acc: 0.7570
Epoch 38/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.2995e-04 - acc: 0.8154Epoch 00038: val_loss did not improve
1712/1712 [==============================] - 2s 1ms/step - loss: 8.3103e-04 - acc: 0.8166 - val_loss: 0.0013 - val_acc: 0.7664
Epoch 39/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.3788e-04 - acc: 0.8184Epoch 00039: val_loss did not improve
1712/1712 [==============================] - 2s 1ms/step - loss: 8.3888e-04 - acc: 0.8183 - val_loss: 0.0013 - val_acc: 0.7593
Epoch 40/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.1628e-04 - acc: 0.8184Epoch 00040: val_loss did not improve
1712/1712 [==============================] - 2s 1ms/step - loss: 8.1458e-04 - acc: 0.8172 - val_loss: 0.0013 - val_acc: 0.7500
Epoch 41/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.3867e-04 - acc: 0.8007Epoch 00041: val_loss did not improve
1712/1712 [==============================] - 2s 1ms/step - loss: 8.3817e-04 - acc: 0.8014 - val_loss: 0.0013 - val_acc: 0.7547
Epoch 42/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.3599e-04 - acc: 0.8290Epoch 00042: val_loss did not improve
1712/1712 [==============================] - 2s 1ms/step - loss: 8.3583e-04 - acc: 0.8283 - val_loss: 0.0014 - val_acc: 0.7687
Epoch 43/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.6192e-04 - acc: 0.8013Epoch 00043: val_loss did not improve
1712/1712 [==============================] - 2s 1ms/step - loss: 8.5971e-04 - acc: 0.8020 - val_loss: 0.0013 - val_acc: 0.7687
Epoch 44/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.6836e-04 - acc: 0.8054Epoch 00044: val_loss did not improve
1712/1712 [==============================] - 2s 1ms/step - loss: 8.6933e-04 - acc: 0.8055 - val_loss: 0.0014 - val_acc: 0.7430
Epoch 45/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.6132e-04 - acc: 0.8025Epoch 00045: val_loss did not improve
1712/1712 [==============================] - 2s 1ms/step - loss: 8.5954e-04 - acc: 0.8020 - val_loss: 0.0013 - val_acc: 0.7523
Epoch 46/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.2890e-04 - acc: 0.8054Epoch 00046: val_loss did not improve
1712/1712 [==============================] - 2s 1ms/step - loss: 8.2888e-04 - acc: 0.8061 - val_loss: 0.0013 - val_acc: 0.7523
Epoch 47/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.1458e-04 - acc: 0.8154Epoch 00047: val_loss did not improve
1712/1712 [==============================] - 2s 1ms/step - loss: 8.1556e-04 - acc: 0.8137 - val_loss: 0.0013 - val_acc: 0.7523
Epoch 48/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.3282e-04 - acc: 0.8278Epoch 00048: val_loss did not improve
1712/1712 [==============================] - 2s 1ms/step - loss: 8.3193e-04 - acc: 0.8271 - val_loss: 0.0013 - val_acc: 0.7757
Epoch 49/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.4104e-04 - acc: 0.8172Epoch 00049: val_loss did not improve
1712/1712 [==============================] - 2s 1ms/step - loss: 8.4064e-04 - acc: 0.8160 - val_loss: 0.0014 - val_acc: 0.7593
Epoch 50/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.4970e-04 - acc: 0.8048Epoch 00050: val_loss did not improve
1712/1712 [==============================] - 2s 1ms/step - loss: 8.5042e-04 - acc: 0.8043 - val_loss: 0.0013 - val_acc: 0.7710
Epoch 51/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.5730e-04 - acc: 0.8166Epoch 00051: val_loss did not improve
1712/1712 [==============================] - 2s 1ms/step - loss: 8.5688e-04 - acc: 0.8160 - val_loss: 0.0014 - val_acc: 0.7570
Epoch 52/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.2532e-04 - acc: 0.8237Epoch 00052: val_loss improved from 0.00131 to 0.00130, saving model to weights/modelweights_Adam.h5
1712/1712 [==============================] - 2s 1ms/step - loss: 8.2624e-04 - acc: 0.8218 - val_loss: 0.0013 - val_acc: 0.7640
Epoch 53/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.5695e-04 - acc: 0.8160Epoch 00053: val_loss did not improve
1712/1712 [==============================] - 2s 1ms/step - loss: 8.5662e-04 - acc: 0.8172 - val_loss: 0.0013 - val_acc: 0.7593
Epoch 54/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.4824e-04 - acc: 0.8184Epoch 00054: val_loss did not improve
1712/1712 [==============================] - 2s 1ms/step - loss: 8.4886e-04 - acc: 0.8166 - val_loss: 0.0013 - val_acc: 0.7640
Epoch 55/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.2031e-04 - acc: 0.8149Epoch 00055: val_loss did not improve
1712/1712 [==============================] - 2s 1ms/step - loss: 8.2083e-04 - acc: 0.8137 - val_loss: 0.0013 - val_acc: 0.7710
Epoch 56/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.5524e-04 - acc: 0.8096Epoch 00056: val_loss did not improve
1712/1712 [==============================] - 2s 1ms/step - loss: 8.5392e-04 - acc: 0.8096 - val_loss: 0.0014 - val_acc: 0.7617
Epoch 57/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.4085e-04 - acc: 0.8178Epoch 00057: val_loss did not improve
1712/1712 [==============================] - 2s 1ms/step - loss: 8.3999e-04 - acc: 0.8183 - val_loss: 0.0013 - val_acc: 0.7547
Epoch 58/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.2377e-04 - acc: 0.8048Epoch 00058: val_loss did not improve
1712/1712 [==============================] - 2s 1ms/step - loss: 8.2814e-04 - acc: 0.8049 - val_loss: 0.0014 - val_acc: 0.7640
Epoch 59/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.4712e-04 - acc: 0.8160Epoch 00059: val_loss did not improve
1712/1712 [==============================] - 2s 1ms/step - loss: 8.4701e-04 - acc: 0.8166 - val_loss: 0.0014 - val_acc: 0.7547
Epoch 60/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.3040e-04 - acc: 0.8125Epoch 00060: val_loss did not improve
1712/1712 [==============================] - 2s 1ms/step - loss: 8.3316e-04 - acc: 0.8125 - val_loss: 0.0013 - val_acc: 0.7593
Epoch 61/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.4252e-04 - acc: 0.8072Epoch 00061: val_loss did not improve
1712/1712 [==============================] - 2s 1ms/step - loss: 8.4101e-04 - acc: 0.8084 - val_loss: 0.0013 - val_acc: 0.7547
Epoch 62/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.4153e-04 - acc: 0.8090Epoch 00062: val_loss did not improve
1712/1712 [==============================] - 2s 1ms/step - loss: 8.4188e-04 - acc: 0.8090 - val_loss: 0.0013 - val_acc: 0.7453
Epoch 63/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.4204e-04 - acc: 0.8072Epoch 00063: val_loss did not improve
1712/1712 [==============================] - 2s 1ms/step - loss: 8.4126e-04 - acc: 0.8072 - val_loss: 0.0013 - val_acc: 0.7430
Epoch 64/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.2880e-04 - acc: 0.8066Epoch 00064: val_loss did not improve
1712/1712 [==============================] - 2s 1ms/step - loss: 8.2785e-04 - acc: 0.8072 - val_loss: 0.0014 - val_acc: 0.7593
Epoch 65/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.3577e-04 - acc: 0.8196Epoch 00065: val_loss did not improve
1712/1712 [==============================] - 2s 1ms/step - loss: 8.3521e-04 - acc: 0.8201 - val_loss: 0.0013 - val_acc: 0.7850
Epoch 66/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.2503e-04 - acc: 0.8172Epoch 00066: val_loss did not improve
1712/1712 [==============================] - 2s 1ms/step - loss: 8.2606e-04 - acc: 0.8172 - val_loss: 0.0013 - val_acc: 0.7710
Epoch 67/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.1812e-04 - acc: 0.8119Epoch 00067: val_loss did not improve
1712/1712 [==============================] - 2s 1ms/step - loss: 8.1864e-04 - acc: 0.8131 - val_loss: 0.0013 - val_acc: 0.7710
Epoch 68/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.3828e-04 - acc: 0.8131Epoch 00068: val_loss did not improve
1712/1712 [==============================] - 2s 1ms/step - loss: 8.3627e-04 - acc: 0.8131 - val_loss: 0.0014 - val_acc: 0.7640
Epoch 69/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.3696e-04 - acc: 0.8101Epoch 00069: val_loss did not improve
1712/1712 [==============================] - 2s 1ms/step - loss: 8.3553e-04 - acc: 0.8096 - val_loss: 0.0014 - val_acc: 0.7593
Epoch 70/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.4454e-04 - acc: 0.8184Epoch 00070: val_loss did not improve
1712/1712 [==============================] - 2s 1ms/step - loss: 8.4621e-04 - acc: 0.8172 - val_loss: 0.0013 - val_acc: 0.7547
Epoch 71/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.4459e-04 - acc: 0.8125Epoch 00071: val_loss did not improve
1712/1712 [==============================] - 2s 1ms/step - loss: 8.4580e-04 - acc: 0.8125 - val_loss: 0.0013 - val_acc: 0.7617
Epoch 72/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.2672e-04 - acc: 0.8166Epoch 00072: val_loss did not improve
1712/1712 [==============================] - 2s 1ms/step - loss: 8.2893e-04 - acc: 0.8160 - val_loss: 0.0014 - val_acc: 0.7500
Epoch 73/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.4880e-04 - acc: 0.7995Epoch 00073: val_loss did not improve
1712/1712 [==============================] - 2s 1ms/step - loss: 8.4906e-04 - acc: 0.7991 - val_loss: 0.0013 - val_acc: 0.7570
Epoch 74/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.6458e-04 - acc: 0.8054Epoch 00074: val_loss did not improve
1712/1712 [==============================] - 2s 1ms/step - loss: 8.6502e-04 - acc: 0.8055 - val_loss: 0.0014 - val_acc: 0.7570
Epoch 75/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.2040e-04 - acc: 0.8149Epoch 00075: val_loss did not improve
1712/1712 [==============================] - 2s 1ms/step - loss: 8.2091e-04 - acc: 0.8154 - val_loss: 0.0013 - val_acc: 0.7664
Epoch 76/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.4046e-04 - acc: 0.8143Epoch 00076: val_loss did not improve
1712/1712 [==============================] - 2s 1ms/step - loss: 8.3859e-04 - acc: 0.8154 - val_loss: 0.0013 - val_acc: 0.7617
Epoch 77/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.3856e-04 - acc: 0.8019Epoch 00077: val_loss did not improve
1712/1712 [==============================] - 2s 1ms/step - loss: 8.3946e-04 - acc: 0.8008 - val_loss: 0.0013 - val_acc: 0.7687
Epoch 78/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.3274e-04 - acc: 0.8208Epoch 00078: val_loss did not improve
1712/1712 [==============================] - 2s 1ms/step - loss: 8.3226e-04 - acc: 0.8213 - val_loss: 0.0013 - val_acc: 0.7617
Epoch 79/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.3075e-04 - acc: 0.8119Epoch 00079: val_loss did not improve
1712/1712 [==============================] - 2s 1ms/step - loss: 8.2981e-04 - acc: 0.8113 - val_loss: 0.0013 - val_acc: 0.7570
Epoch 80/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.4496e-04 - acc: 0.8107Epoch 00080: val_loss did not improve
1712/1712 [==============================] - 2s 1ms/step - loss: 8.4424e-04 - acc: 0.8096 - val_loss: 0.0013 - val_acc: 0.7523
Epoch 81/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.5587e-04 - acc: 0.8096Epoch 00081: val_loss did not improve
1712/1712 [==============================] - 2s 1ms/step - loss: 8.5638e-04 - acc: 0.8102 - val_loss: 0.0013 - val_acc: 0.7687
Epoch 82/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.3413e-04 - acc: 0.8066Epoch 00082: val_loss did not improve
1712/1712 [==============================] - 2s 1ms/step - loss: 8.3320e-04 - acc: 0.8061 - val_loss: 0.0013 - val_acc: 0.7664
Epoch 83/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.4360e-04 - acc: 0.8060Epoch 00083: val_loss did not improve
1712/1712 [==============================] - 2s 1ms/step - loss: 8.4101e-04 - acc: 0.8072 - val_loss: 0.0013 - val_acc: 0.7687
Epoch 84/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.6075e-04 - acc: 0.8125Epoch 00084: val_loss did not improve
1712/1712 [==============================] - 2s 1ms/step - loss: 8.6076e-04 - acc: 0.8125 - val_loss: 0.0014 - val_acc: 0.7664
Epoch 85/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.5529e-04 - acc: 0.8066Epoch 00085: val_loss did not improve
1712/1712 [==============================] - 2s 1ms/step - loss: 8.5654e-04 - acc: 0.8072 - val_loss: 0.0014 - val_acc: 0.7500
Epoch 86/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.2886e-04 - acc: 0.8001Epoch 00086: val_loss did not improve
1712/1712 [==============================] - 2s 1ms/step - loss: 8.3017e-04 - acc: 0.7991 - val_loss: 0.0013 - val_acc: 0.7780
Epoch 87/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.2672e-04 - acc: 0.8090Epoch 00087: val_loss did not improve
1712/1712 [==============================] - 2s 1ms/step - loss: 8.2708e-04 - acc: 0.8078 - val_loss: 0.0013 - val_acc: 0.7617
Epoch 88/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.1573e-04 - acc: 0.8249Epoch 00088: val_loss improved from 0.00130 to 0.00130, saving model to weights/modelweights_Adam.h5
1712/1712 [==============================] - 2s 1ms/step - loss: 8.1785e-04 - acc: 0.8230 - val_loss: 0.0013 - val_acc: 0.7780
Epoch 89/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.2810e-04 - acc: 0.8202Epoch 00089: val_loss did not improve
1712/1712 [==============================] - 2s 1ms/step - loss: 8.2803e-04 - acc: 0.8201 - val_loss: 0.0014 - val_acc: 0.7547
Epoch 90/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.2325e-04 - acc: 0.8084Epoch 00090: val_loss did not improve
1712/1712 [==============================] - 2s 1ms/step - loss: 8.2200e-04 - acc: 0.8090 - val_loss: 0.0013 - val_acc: 0.7617
Epoch 91/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.1610e-04 - acc: 0.7960Epoch 00091: val_loss did not improve
1712/1712 [==============================] - 2s 1ms/step - loss: 8.1672e-04 - acc: 0.7944 - val_loss: 0.0013 - val_acc: 0.7640
Epoch 92/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.4020e-04 - acc: 0.8113Epoch 00092: val_loss did not improve
1712/1712 [==============================] - 2s 1ms/step - loss: 8.3980e-04 - acc: 0.8107 - val_loss: 0.0013 - val_acc: 0.7640
Epoch 93/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.6198e-04 - acc: 0.8119Epoch 00093: val_loss did not improve
1712/1712 [==============================] - 2s 1ms/step - loss: 8.5938e-04 - acc: 0.8119 - val_loss: 0.0013 - val_acc: 0.7593
Epoch 94/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.1834e-04 - acc: 0.8190Epoch 00094: val_loss did not improve
1712/1712 [==============================] - 2s 1ms/step - loss: 8.1874e-04 - acc: 0.8195 - val_loss: 0.0014 - val_acc: 0.7640
Epoch 95/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.2172e-04 - acc: 0.8219Epoch 00095: val_loss did not improve
1712/1712 [==============================] - 2s 1ms/step - loss: 8.2118e-04 - acc: 0.8218 - val_loss: 0.0013 - val_acc: 0.7617
Epoch 96/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.1460e-04 - acc: 0.8137Epoch 00096: val_loss did not improve
1712/1712 [==============================] - 2s 1ms/step - loss: 8.1479e-04 - acc: 0.8137 - val_loss: 0.0013 - val_acc: 0.7757
Epoch 97/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.3432e-04 - acc: 0.8154Epoch 00097: val_loss did not improve
1712/1712 [==============================] - 2s 1ms/step - loss: 8.3495e-04 - acc: 0.8137 - val_loss: 0.0013 - val_acc: 0.7757
Epoch 98/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.1440e-04 - acc: 0.8078Epoch 00098: val_loss did not improve
1712/1712 [==============================] - 2s 1ms/step - loss: 8.1320e-04 - acc: 0.8084 - val_loss: 0.0013 - val_acc: 0.7664
Epoch 99/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.2109e-04 - acc: 0.8084Epoch 00099: val_loss did not improve
1712/1712 [==============================] - 2s 1ms/step - loss: 8.2178e-04 - acc: 0.8096 - val_loss: 0.0013 - val_acc: 0.7664
Epoch 100/100
1696/1712 [============================>.] - ETA: 0s - loss: 8.2724e-04 - acc: 0.8154Epoch 00100: val_loss did not improve
1712/1712 [==============================] - 2s 1ms/step - loss: 8.2619e-04 - acc: 0.8143 - val_loss: 0.0013 - val_acc: 0.7757

---

Step 7: Visualize the Loss and Test Predictions

(IMPLEMENTATION) Answer a few questions and visualize the loss

Question 1: Outline the steps you took to get to your final neural network architecture and your reasoning at each step.

Answer: I started with sort of what I had in the dog project but then tried to tweak the hyper parameters a bit. It didn't change much to the end result, it's around a tenth of a percent, so that's fairly okay I think. The accuracy might have gotten a little higher with more data and more epochs. I used dropout to ensure that all neurons train.

Question 2: Defend your choice of optimizer. Which optimizers did you test, and how did you determine which worked best?

Answer: I tested all of them. SGD is the best in my case but they are all quite similar in performance, only Nadam does not score so well.

Use the code cell below to plot the training and validation loss of your neural network. You may find this resource useful.

## Visualize the training and validation loss of your neural network

legends = []
fig = plt.figure(figsize=(15,15))
ax1 = fig.add_subplot(221)
ax2 = fig.add_subplot(222)
ax3 = fig.add_subplot(223)

for optimizer in optimizers:
    
    #if optimizer == 'Nadam':
        #continue
    
    accuracy = model_histories[optimizer].history['val_acc']
    loss = model_histories[optimizer].history['loss']
    val_loss = model_histories[optimizer].history['val_loss']
    
    print("Optimizer:".ljust(20), optimizer, "\t Loss: ", loss[-1], "\t Validation loss: ", val_loss[-1])
    
    ax1.plot(loss, label = "Training Loss")
    ax2.plot(val_loss, label = "Validation Loss")
    ax3.plot(accuracy, label = "Training Accuracy")
    legends.append(optimizer)

ax1.set_title('Training loss')
ax1.set_ylabel('Loss')
ax1.set_xlabel('Epochs')
ax1.legend(legends, loc='upper right')

ax2.set_title('Validation Loss')
ax2.set_ylabel('Loss')
ax2.set_xlabel('Epochs')
ax2.legend(legends, loc='upper right')

ax3.set_title('Validation accuracy')
ax3.set_ylabel('Accuracy')
ax3.set_xlabel('Epochs')
ax3.legend(legends, loc='lower right')

Optimizer:           Nadam 	 Loss:  0.00101386834584 	 Validation loss:  0.00144490605507
Optimizer:           Adadelta 	 Loss:  0.000869528318116 	 Validation loss:  0.00131339007754
Optimizer:           Adamax 	 Loss:  0.000827893025076 	 Validation loss:  0.00132267638414
Optimizer:           Adagrad 	 Loss:  0.000792255129004 	 Validation loss:  0.00132557710798
Optimizer:           RMSprop 	 Loss:  0.000880617575468 	 Validation loss:  0.00134157168757
Optimizer:           SGD 	 Loss:  0.000790864636112 	 Validation loss:  0.00130201196167
Optimizer:           Adam 	 Loss:  0.000826187833826 	 Validation loss:  0.0013313752536

<matplotlib.legend.Legend at 0x7ff08356ccc0>

Question 3: Do you notice any evidence of overfitting or underfitting in the above plot? If so, what steps have you taken to improve your model? Note that slight overfitting or underfitting will not hurt your chances of a successful submission, as long as you have attempted some solutions towards improving your model (such as regularization, dropout, increased/decreased number of layers, etc).

Answer: There is some slight overfitting, I have done dropout and used 5 convolutional layers with pooling layers to make sure that it doesn't overfit the data too much. In this project I did not use batch normalization as I don't think it would have made a big difference. I did this in other projects and only in case of gross overfitting was this required.

Visualize a Subset of the Test Predictions

Execute the code cell below to visualize your model's predicted keypoints on a subset of the testing images.

y_test = model.predict(X_test)
fig = plt.figure(figsize=(20,20))
fig.subplots_adjust(left=0, right=1, bottom=0, top=1, hspace=0.05, wspace=0.05)
for i in range(9):
    ax = fig.add_subplot(3, 3, i + 1, xticks=[], yticks=[])
    plot_data(X_test[i], y_test[i], ax)

---

Step 8: Complete the pipeline

With the work you did in Sections 1 and 2 of this notebook, along with your freshly trained facial keypoint detector, you can now complete the full pipeline. That is given a color image containing a person or persons you can now

• Detect the faces in this image automatically using OpenCV
• Predict the facial keypoints in each face detected in the image
• Paint predicted keypoints on each face detected

In this Subsection you will do just this!

(IMPLEMENTATION) Facial Keypoints Detector

Use the OpenCV face detection functionality you built in previous Sections to expand the functionality of your keypoints detector to color images with arbitrary size. Your function should perform the following steps

1. Accept a color image.
2. Convert the image to grayscale.
3. Detect and crop the face contained in the image.
4. Locate the facial keypoints in the cropped image.
5. Overlay the facial keypoints in the original (color, uncropped) image.

Note: step 4 can be the trickiest because remember your convolutional network is only trained to detect facial keypoints in $96 \times 96$ grayscale images where each pixel was normalized to lie in the interval $[0,1]$, and remember that each facial keypoint was normalized during training to the interval $[-1,1]$. This means - practically speaking - to paint detected keypoints onto a test face you need to perform this same pre-processing to your candidate face - that is after detecting it you should resize it to $96 \times 96$ and normalize its values before feeding it into your facial keypoint detector. To be shown correctly on the original image the output keypoints from your detector then need to be shifted and re-normalized from the interval $[-1,1]$ to the width and height of your detected face.

When complete you should be able to produce example images like the one below

# Load in color image for face detection
image = cv2.imread('images/obamas4.jpg')


# Convert the image to RGB colorspace
image_copy = cv2.cvtColor(image, cv2.COLOR_BGR2RGB)


# plot our image
fig = plt.figure(figsize = (9,9))
ax1 = fig.add_subplot(111)
ax1.set_xticks([])
ax1.set_yticks([])
ax1.set_title('image copy')
ax1.imshow(image_copy)

<matplotlib.image.AxesImage at 0x7ff0822114e0>

def detect_just_faces(image, sens=4):
    # Convert the RGB  image to grayscale
    gray_noise = cv2.cvtColor(image, cv2.COLOR_RGB2GRAY)

    # Extract the pre-trained face detector from an xml file
    face_cascade = cv2.CascadeClassifier('detector_architectures/haarcascade_frontalface_default.xml')

    # Detect the faces in image
    faces = face_cascade.detectMultiScale(gray_noise, sens, 6)

    # Print the number of faces detected in the image
    print('Number of faces detected:', len(faces))

    return faces

def image_with_box_faces(image, faces):
    # Make a copy of the orginal image to draw face detections on
    image_with_detections = np.copy(image)

    # Get the bounding box for each detected face
    for (x,y,w,h) in faces:
        # Add a red bounding box to the detections image
        cv2.rectangle(image_with_detections, (x,y), (x+w,y+h), (255,0,0), 3)
    
    return image_with_detections

def crop_faces(image, faces):
    for (x,y,w,h) in faces:
        face = image[y:y+h, x:x+w]
        resized_image = cv2.resize(face, (96, 96)) 
        return resized_image

def normalize_image(image_unnormalized):
    image_normalized = cv2.cvtColor(image_unnormalized, cv2.COLOR_RGB2GRAY) / 255
    image_normalized = np.expand_dims(np.expand_dims(image_normalized, axis=-1), axis=0)
    return image_normalized

def get_keypoints(image_normalized):
    return model.predict(image_normalized)[0]

def denormalize(face, keypoints):
    (x, y, w, h) = face
    keypoints_x = keypoints[0::2] * (w / 2) + (w / 2) + x
    keypoints_y = keypoints[1::2] * (h / 2) + (h / 2) + y
    return list(zip(keypoints_x, keypoints_y))

def get_image_with_keypoints(image, keypoints):
    for (x, y) in keypoints:
        cv2.circle(image, (x,y), 3, (0,255,0), -1)
    return image

### TODO: Use the face detection code we saw in Section 1 with your trained conv-net 
## TODO : Paint the predicted keypoints on the test image

# Accept a color image.
# Convert the image to grayscale.
faces_obamas = detect_just_faces(image_copy, 1.25)
image_obamas = image_with_box_faces(image_copy, faces_obamas)
keypoints_list = []

for face_obamas in faces_obamas:
    print(face_obamas)
    
    cropped_faces = crop_faces(image_copy, faces_obamas)
    normalized_faces = normalize_image(cropped_face)
    keypoints_predicted = get_keypoints(normalized_faces)
    keypoints = denormalize(face_obamas, keypoints_predicted)
    image_keypoints = get_image_with_keypoints(image_obamas, keypoints)

fig = plt.figure(figsize = (9,9))
ax1 = fig.add_subplot(111)
ax1.set_xticks([])
ax1.set_yticks([])
ax1.set_title('Image with detections')
ax1.imshow(image_keypoints)

Number of faces detected: 2
[179  74 175 175]
[366 141 166 166]

<matplotlib.image.AxesImage at 0x7ff082788048>

(Optional) Further Directions - add a filter using facial keypoints to your laptop camera

Now you can add facial keypoint detection to your laptop camera - as illustrated in the gif below.

The next Python cell contains the basic laptop video camera function used in the previous optional video exercises. Combine it with the functionality you developed for keypoint detection and marking in the previous exercise and you should be good to go!

import cv2
import time 
from keras.models import load_model
def laptop_camera_go():
    # Create instance of video capturer
    cv2.namedWindow("face detection activated")
    vc = cv2.VideoCapture(0)

    # Try to get the first frame
    if vc.isOpened(): 
        rval, frame = vc.read()
    else:
        rval = False
    
    # keep video stream open
    while rval:
        # plot image from camera with detections marked
        cv2.imshow("face detection activated", frame)
        
        # exit functionality - press any key to exit laptop video
        key = cv2.waitKey(20)
        if key > 0: # exit by pressing any key
            # destroy windows
            cv2.destroyAllWindows()
            
            # hack from stack overflow for making sure window closes on osx --> https://stackoverflow.com/questions/6116564/destroywindow-does-not-close-window-on-mac-using-python-and-opencv
            for i in range (1,5):
                cv2.waitKey(1)
            return
        
        # read next frame
        time.sleep(0.05)             # control framerate for computation - default 20 frames per sec
        rval, frame = vc.read()  

# Run your keypoint face painter
laptop_camera_go()

(Optional) Further Directions - add a filter using facial keypoints

Using your freshly minted facial keypoint detector pipeline you can now do things like add fun filters to a person's face automatically. In this optional exercise you can play around with adding sunglasses automatically to each individual's face in an image as shown in a demonstration image below.

To produce this effect an image of a pair of sunglasses shown in the Python cell below.

# Load in sunglasses image - note the usage of the special option
# cv2.IMREAD_UNCHANGED, this option is used because the sunglasses 
# image has a 4th channel that allows us to control how transparent each pixel in the image is
sunglasses = cv2.imread("images/sunglasses_4.png", cv2.IMREAD_UNCHANGED)

# Plot the image
fig = plt.figure(figsize = (6,6))
ax1 = fig.add_subplot(111)
ax1.set_xticks([])
ax1.set_yticks([])
ax1.imshow(sunglasses)
ax1.axis('off');

This image is placed over each individual's face using the detected eye points to determine the location of the sunglasses, and eyebrow points to determine the size that the sunglasses should be for each person (one could also use the nose point to determine this).

Notice that this image actually has 4 channels, not just 3.

# Print out the shape of the sunglasses image
print ('The sunglasses image has shape: ' + str(np.shape(sunglasses)))

It has the usual red, blue, and green channels any color image has, with the 4th channel representing the transparency level of each pixel in the image. Here's how the transparency channel works: the lower the value, the more transparent the pixel will become. The lower bound (completely transparent) is zero here, so any pixels set to 0 will not be seen.

This is how we can place this image of sunglasses on someone's face and still see the area around of their face where the sunglasses lie - because these pixels in the sunglasses image have been made completely transparent.

Lets check out the alpha channel of our sunglasses image in the next Python cell. Note because many of the pixels near the boundary are transparent we'll need to explicitly print out non-zero values if we want to see them.

# Print out the sunglasses transparency (alpha) channel
alpha_channel = sunglasses[:,:,3]
print ('the alpha channel here looks like')
print (alpha_channel)

# Just to double check that there are indeed non-zero values
# Let's find and print out every value greater than zero
values = np.where(alpha_channel != 0)
print ('\n the non-zero values of the alpha channel look like')
print (values)

This means that when we place this sunglasses image on top of another image, we can use the transparency channel as a filter to tell us which pixels to overlay on a new image (only the non-transparent ones with values greater than zero).

One last thing: it's helpful to understand which keypoint belongs to the eyes, mouth, etc. So, in the image below, we also display the index of each facial keypoint directly on the image so that you can tell which keypoints are for the eyes, eyebrows, etc.

With this information, you're well on your way to completing this filtering task! See if you can place the sunglasses automatically on the individuals in the image loaded in / shown in the next Python cell.

# Load in color image for face detection
image = cv2.imread('images/obamas4.jpg')

# Convert the image to RGB colorspace
image = cv2.cvtColor(image, cv2.COLOR_BGR2RGB)


# Plot the image
fig = plt.figure(figsize = (8,8))
ax1 = fig.add_subplot(111)
ax1.set_xticks([])
ax1.set_yticks([])
ax1.set_title('Original Image')
ax1.imshow(image)

## (Optional) TODO: Use the face detection code we saw in Section 1 with your trained conv-net to put
## sunglasses on the individuals in our test image

(Optional) Further Directions - add a filter using facial keypoints to your laptop camera

Now you can add the sunglasses filter to your laptop camera - as illustrated in the gif below.

The next Python cell contains the basic laptop video camera function used in the previous optional video exercises. Combine it with the functionality you developed for adding sunglasses to someone's face in the previous optional exercise and you should be good to go!

import cv2
import time 
from keras.models import load_model
import numpy as np

def laptop_camera_go():
    # Create instance of video capturer
    cv2.namedWindow("face detection activated")
    vc = cv2.VideoCapture(0)

    # try to get the first frame
    if vc.isOpened(): 
        rval, frame = vc.read()
    else:
        rval = False
    
    # Keep video stream open
    while rval:
        # Plot image from camera with detections marked
        cv2.imshow("face detection activated", frame)
        
        # Exit functionality - press any key to exit laptop video
        key = cv2.waitKey(20)
        if key > 0: # exit by pressing any key
            # Destroy windows 
            cv2.destroyAllWindows()
            
            for i in range (1,5):
                cv2.waitKey(1)
            return
        
        # Read next frame
        time.sleep(0.05)             # control framerate for computation - default 20 frames per sec
        rval, frame = vc.read()    
        

# Load facial landmark detector model
model = load_model('my_model.h5')

# Run sunglasses painter
laptop_camera_go()